Microsoft Word - Vasantha-book7.doc

w. b. vasantha kandasamy

SMARANDACHE

NON-ASSOCIATIVE

RINGS

AMERICAN RESEARCH PRESS

REHOBOTH, NM

2002

Smarandache

Non-associative rings

W. B. Vasantha Kandasamy

Department of Mathematics

Indian Institute of Technology, Madras

Chennai – 600036, India

American Research Press

Rehoboth, NM

2002

This book can be ordered in a paper bound reprint from:

Books

Demand

ProQuest

Information

Learning

(University

Microfilm

International)

300

Zeeb

Road

P.O. Box 1346, Ann Arbor

48106-1346,

USA

Tel.:

1-800-521-0600

(Customer

Service)

http://wwwlib.umi.com/bod/

and online from:

        Publishing Online, Co. (Seattle, Washington State)
        at: http://PublishingOnline.com

This book has been peer reviewed and recommended for publication by:

Dr. Mihaly Bencze, Editor, Octogon, 6, Hatmanului Street, 2212 Sacele 3, Jud Brasov, Romania.
Dr. Andrei Kelarev, Dept. of Mathematics, Univ. of Tasmania, Hobart, Tasmania 7001, Australia.
Dr. A. R. T. Solarin, Dept. of Mathematics, Obafemi Awolowo University, (formerly University of
Ife), Ile-Ife, Nigeria.

Copyright 2002 by American Research Press and W. B. Vasantha Kandasamy
Rehoboth, Box 141
NM 87322, USA

Many books can be downloaded from:
http://www.gallup.unm.edu/~smarandache/eBooks-otherformats.htm

ISBN:  1-931233-69-1

Standard Address Number: 297-5092
Printed in the United States of America

CONTENTS

Preface

1. BASIC CONCEPTS

1.1 Basics of vector space and bilinear forms

1.2 Smarandache vector spaces

1.3 Basic definitions of other algebraic structures

2. LOOP-RINGS AND SMARANDACHE LOOP-RINGS

2.1 Introduction of loops and Smarandache loops

2.2 Properties of loop-rings and introduction to Smarandache loop-rings

2.3 Smarandache elements in loop rings

2.4 Smarandache substructures in loop rings

2.5 General properties of Smarandache loop rings and loop rings

3. GROUPOID RINGS AND SMARANDACHE GROUPOID RINGS

3.1 Groupoids and Smarandache Groupoids

3.2 Groupoid rings and Smarandache Groupoid rings

3.3 Smarandache special elements in Groupoid rings

3.4 Smarandache substructures in Groupoid rings

3.5 Special properties in Groupoid rings and Smarandache groupoid rings

4. LIE ALGEBRAS AND SMARANDACHE LIE ALGEBRAS

4.1 Basic properties of Lie algebras

4.2 Smarandache Lie algebras and its basic properties

4.3 Substructures in Smarandache Lie algebras

4.4 Special properties in Smarandache Lie algebras

103

4.5 Some new notions on Smarandache Lie algebras

106

5. JORDAN ALGEBRAS AND SMARANDACHE JORDAN
ALGEBRAS

5.1 Basic properties of Jordan algebras

109

5.2 Smarandache Jordan algebras and its basic properties

113

6. SUGGESTIONS FOR FUTURE STUDY

6.1 Smarandache groupoid rings

119

6.2 Smarandache loop rings

120

6.3 Lie algebras and Smarandache Lie algebras

121

6.4 Smarandache Jordan algebras

122

6.5 Other non-associative rings and algebras and

their Smarandache analogue

123

7. SUGGESTED PROBLEMS

127

References

139

Index

145

PREFACE

An associative ring is just realized or built using reals or complex; finite or infinite
by defining two binary operations on it. But on the contrary when we want to define
or study or even introduce a non-associative ring we need two separate algebraic
structures say a commutative ring with 1 (or a field) together with a loop or a
groupoid or a vector space or a linear algebra. The two non-associative well-known
algebras viz. Lie algebras and Jordan algebras are mainly built using a vector space
over a field satisfying special identities called the Jacobi identity and Jordan identity
respectively. Study of these algebras started as early as 1940s. Hence the study of
non-associative algebras or even non-associative rings boils down to the study of
properties of vector spaces or linear algebras over fields.

But study of non-associative algebras using loops, that is loop algebras (or loop
rings) was at its peak only in the 1980s. But till date there is no separate book on
loop rings. So in this book we have given more importance to the study of loop rings
and above all Smarandache loop rings. Further in my opinion the deeper study of
loop rings will certainly branch off into at least five special type of algebras like
Moufang algebras, Bol algebras, Bruck algebras, Alternative algebras and P-
algebras. The author does not deny there are several research papers on these five
special algebras but each of these algebras can be developed like Lie algebras and
Jordan algebras.

Another generalized class of non-associative algebras are from groupoid rings. The
class of loop rings are strictly contained in the class of groupoid rings. Several
important study in the direction of Smarandache notions has been done in this book.
Groupoid rings are a very new concept, till date only very few research papers exist
in the field of Groupoid rings. The notion of Smarandache Lie algebras and
Smarandache Jordan algebras are introduced. There are several informative and
exhaustive books on Lie algebras and Jordan algebras. Further as the study of these
two algebras heavily exploits the concepts of vector spaces and matrix theory we do
not venture to approach Smarandache Lie algebras or Smarandache Jordan algebras
in this direction. But in this book we approach Smarandache Lie algebras and
Smarandache Jordan algebras as a study in par with associative ringsfor such a study
has not yet been approached by any researcher in this way. This book is distinct and
different from other non-associative ring theory books as the properties of
associative rings are incorporated and studied even in Lie rings and Jordan rings, an
approach which is not traditional.

This book has seven chapters. First chapter is not only introductory but also devotes
an entire section to introduce the basic concepts in Smarandache vector spaces.
Chapter two studies about Smarandache loop rings and in chapter three, the
Smarandache groupoid rings is introduced and analysed. Chapter four is devoted to
the introduction of several new concepts in Lie algebras and Smarandache Lie
algebras. Chapter five defines Smarandache Jordan algebras and gives several new
notions so far not studied about in Jordan algebras and Smarandache Jordan algebras.
Overall, we have introduced nearly 160 Smarandache notions relating to non-
associative rings. The sixth chapter gives suggestions for future study. The last
chapter gives about 150 problems which will help any researcher.

I first deeply acknowledge my indebtness to Dr. Minh Perez of American Research
Press, whose constant mental support and encouragement made me write this book.

I thank my husband Dr. Kandasamy and my daughters Meena and Kama, who
tirelessly worked for very long hours, typing and proof-reading this book.

I dedicate this book to Mahatma Jotirao Phule (1827-1890), a veteran social
revolutionary whose progressive ideas and activism – encompassing the triple
concerns of caste, class and gender – heralded a definitely different future for India's
oppressed masses. At the age of twenty, Phule established the first and only school
for the girls of untouchable and oppressed castes, at a period of time in history where
women and the oppressed castes were denied all education. His greatness remains in
his challenging the ruthless gender-based and caste-based domination over
education, a domination which still continues and requires our intensive struggle.

Chapter 1

BASIC CONCEPTS

This chapter on basic concepts has three sections. First section mainly describes the
specific notions of vector space and bilinear forms. In section two we develop the
definition of Smarandache K-vectorial space given in [43]. In fact we develop
several important properties in this direction, which has so far remained as an
undeveloped concept. In the final section we just recall the definition of semigroups,
groupoids, S-semigroups, S-groupoids and loops.

1.1 Basics of Vector space and bilinear forms

In this section we just recall the concept of vector space, bilinear form, symmetric
and skew symmetric forms. As we expect the reader to be well versed with concepts
in algebra we do not recall the definition of fields or rings.

D

EFINITION

1.1.1: A vector space or a linear space consists of the following:

1.  a field K of scalars.
2.  a set V of elements called vectors.
3.  a rule or operation called vector addition which associates with each pair

of vectors

α ,β in V a vector α +β in V, called the sum of α and β in such

a way that

a. addition is commutative

α + β =β + α .

b. addition is associative that is

α + (β + γ) = (α + β) + γ.

c. There is a unique vector 0 in V, called the zero vector

such that

α + 0 = α for all α ∈ V.

d. For each vector

α in V there is a unique vector - α in V

such that

α + (-α) = 0.

4. An operation called scalar multiplication, which associates with each

scalar c in K and a vector

α in V a vector cα in V called the product of c

and

α in such a way That

a. 1.

α = α for all α ∈ V.

b. (a

)

α = a

α).

c. a(

α+β) = a α + aβ.

d. (a

+ a

)

α = a

α +a

α, a

, a

∈ K and α, β ∈ V.

Thus we see a vector space unlike the abstract structures rings or fields is a
composite structure consisting of two algebraic structures called fields and vectors,
which forms an additive abelian group. Here it is pertinent to mention that 'vector'
does not imply it is a vector quantity, only to mention that it is different from the
elements of the field which are termed as scalars.

EFINITION

1.1.2: A subspace of the vector space V is a subset W of V which is itself

a vector space over F with the operations of vector addition and scalar
multiplication on V.

The reader is advised to learn the concept of basis, linear transformation, linear
operator etc. We say a vector space is finite dimensional if there exists a finite set of
linearly independent elements (v

, v

, …, v

) in V which generate V, otherwise it is

infinite dimensional that is there does not exist a finite set of linearly independent
elements in V which can span V.

The following definitions will be used in the later chapters, hence we recall them.

D

EFINITION

1.1.3: A bilinear form on a vector space V (real or complex) is a

mapping [x, y]

→ (x, y) of V × V into the set of (real or complex) numbers such that

(ax + bx', y) = a (x, y) + b (x', y) and

(x, ay + by') = a (x, y) + b (x, y')

for all vectors x, x' y, y' in V and scalars a, b in the field.

D

EFINITION

1.1.4: A bilinear form is skew symmetric if (x, y) = – (y, x) for all

vectors x, y. A Hermitian form on a complex vector space V is a mapping [x, y]

→

(x, y) of V

× V into the set of complex numbers such that (x, y) =

)

(

and (ax +

bx', y) = a (x, y) + b (x', y) so that (x, ay + by') =

)

(

)

(

−

for all vectors x,

x', y, y' and scalars a, b. A bilinear form (x, y) is called non-singular if for every x

∈

V (x

≠ 0) the linear form (x, x

) is not identically 0. Otherwise the bilinear form is

said to be singular. This form is said to be positive definite if (x, x) > 0 for all x

≠ 0;

∈ V.

EFINITION

1.1.5: A subset B of an associative algebra A over a field F is called

weakly closed if for every ordered pair (a, b), a, b

∈ B, there is defined an element

τ(a, b) ∈ F such that ab + τ (a, b) ba ∈ B. We assume the mapping (a, b)→ τ (a,b) is
fixed and write a

× b = ab + τ (a, b ) ba.

A subset

Π of B is called a subsystem if c × d ∈ Π for every c, d ∈ Π, and Π is a left

ideal (ideal) if a

× c ∈ Π (a × c and c × a ∈ Π) for every a ∈ B and c ∈ Π. GL (V) or

gl (v) denotes the group of all non-singular linear transformations of V, this is a
linear algebraic group.

1.2 Smarandache vector spaces

In this section we introduce the concept of Smarandache K-vectoroial space as given
in [43]. We further illustrate with examples. As no one has worked on the
Smarandache properties of Smarandache K-vectorial spaces here we have developed
some of the concepts, which will be very useful in the study of linear structure and
Lie algebra. The notions of Smarandache basis and Smarandache transformations are
defined.

EFINITION

[43]: The Smarandache K-vectorial space (S-K-vectorial space) is

defined to be K-vectorial space (A, +, .) such that a proper subset of A is a K-
algebra (with respect to the same induced operation and another ‘

×’ operation

internal on A where K is a commutative field).

By a proper subset we understand a set included in A, different from the empty set,
from the unit element if any, and from A.

Example 1.2.1: Let V

= F

m×n

be a vector space over Q and V

= Q [x] be a vector

space over Q. Then V = V

× V

, the direct product of the vector spaces over Q. We

see V is a S-Q-vectorial space. For {0}

× V

⊂ V is a proper subset of V is a Q-

algebra with respect to the same induced operation and the ‘

×’ in V

being the

polynomial multiplication. Here K = Q.

Example 1.2.2: V = V

× V

where V

= Q

× Q is a vector space over Q, V

Q[x] is a vector space over Q and V

= {set of all odd degree polynomials with

coefficients from Q}. V is a S-Q-vectorial space and

V = V

× {0} × {0} is a proper

subset of V which is a K-algebra.

Similarly

V = {0}

× V

× {0} is a proper subset which is a Q-algebra over Q. Thus

by this example we illustrate that there can be more than one proper subset which is
a K-algebra, for K = Q in this example. Now we proceed on to define a Smarandache
K-vectorial subspace.

D

EFINITION

1.2.1: Let A be a K-vectorial space. A proper subset X of A is said to be

a Smarandache K-vectorial subspace (S-K-vectorial subspace) of A if X itself is a S-
K-vectorial space.

T

HEOREM

1.2.1: Let A be a K-vectorial space. If A has a S-K-vectorial subspace

then A is S-K-vectorial space.

Proof: Follows from the very definition. Thus we need not say that A is a S-K-
vectorial space for if A has a S-K-vectorial subspace then A itself is a S-K-vectorial
space.

The study of Smarandache basis is really a new concept.

D

EFINITION

1.2.2: Let V be a finite dimensional vector space over a field K. Let B =

, v

, …, v

} be a basis of V. We say B is a Smarandache basis (S-basis) of V if B

has a proper subset say A, A

⊂ B and A ≠ φ; A ≠ B such that A generates a subspace

which is linear algebra over K that is if W is the subspace generated by A then W
must be a K-algebra with the same operation of V.

Thus it is interesting to note that all basis of a vector space need not in general be a
S-basis.

T

HEOREM

1.2.2: Let V be a vector space over the field K. If B is a S-basis then B is a

basis of V.

Proof: Obvious by the very definition of these concepts.

D

EFINITION

1.2.3: Let V be a finite dimensional vector space over a field K. Let B =

, v

, …, v

} be a basis of V . If every proper subset of B generates a linear

algebra over K then we call B a Smarandache strong basis (S-strong basis) for V.

D

EFINITION

1.2.4: Let L be any vector space over the field K. We say L is a

Smarandache finite dimensional vector space (S-finite dimensional vector space) of
K if every S-basis has only finite number of elements in it. It is interesting to note
that if L is a finite dimensional vector space then L is a S-finite dimensional space
provided L has a S-basis.

It can also happen that L need not be a finite dimensional space still L can be a S-
finite dimensional space.

T

HEOREM

1.2.3: Let V be a vector space over the field K. If A = {v

, v

, …, v

} is a

S-strong basis of V then A is a S-basis of V.

Proof: Easily follows from the very definition of these concepts.

T

HEOREM

1.2.4: Let V be a vector space over the field K. If A = {v

, v

, …, v

} is a

S-basis of V. A need not in general be a S-strong basis of V.

Proof: By an example. Let V = Q [x] be the set of all polynomials of degree less than
or equal to 10. V is a vector space over Q. Clearly A = {1, x, x

, …, x

} is a basis of

V. In fact A is a S-basis of V for take B = {1, x

, x

}. Clearly B generates

a linear algebra. But all subsets of A do not form a S-basis of V, so A is not a S-
strong basis of V but only a S-basis of V. Hence the claim.

Now we proceed on to define Smarandache eigen values and Smarandache eigen
vectors of a vector space.

D

EFINITION

1.2.5: Let V be a vector space over the field F and let T be a linear

operator from V to V. T is said to be a Smarandache linear operator (S-linear
operator) on V if V has a S-basis, which is mapped by T onto another S-basis of V.

D

EFINITION

1.2.6: Let T be a S-linear operator defined on the space V. A

characteristic value c in F of T is said to be a Smarandache characteristic value (S-
characteristic value) of T if the characteristic vector of T associated with c generate
a subspace which is a linear algebra, that is the characteristic space associated with
c is a linear algebra. So the eigen vector associated with the S-characteristic values
will be called as Smarandache eigen vectors (S-eigen vectors) or Smarandache
characteristic vectors (S-characteristic vectors).

All other related concepts can be developed. But as this book is on non-associative
ring we may not require all these properties we have just freshly defined the essential
ones for making use of them in later chapters.

1.3 Basic definitions of other algebraic structures

In this section we just define the concept of semigroups, groupoids, S-semigroups, S-
groupoid, S-loops and loops. For in chapters 2 and 3 the non-associative rings are
introduced mainly using these algebraic structures loops and groupoids viz loop rings
and groupoid rings. These rings are non-associative having several properties other
than the properties enjoyed by Lie algebras and Jordan algebras.

D

EFINITION

1.3.1: Let S be a non-empty set on which is defined a binary operation

'.' such that

1. a.b

∈ S for all a, b ∈ S. (closure axiom)

2. a. (b.c) = (a.b).c for all a, b, c

∈ L. (associative law)

(S, .) is called the semigroup. The semigroup in general may not have identity. If the
semigroup S has an element e

∈ S, such that s.e = e.s = s for all s ∈ S, then we call

(S, .) a semigroup with unit or a monoid.

D

EFINITION

1.3.2: Let G be a non-empty set on which is defined a binary operation

‘.’ such that S is closed under ‘.’ But ‘.’ on S is not in general associative, then we
say (S, .) is a groupoid. Thus a groupoid is a non-associative semigroup.

A groupoid may or may not have the identity. If it has identity then we call the
groupoid to be a groupoid with identity. Thus trivially all semigroups are groupoids
and all groupoids in general need not be semigroups.

Example 1.3.1: Let Z

be the set of integers. Define an operation ‘.’ on Z

by a.b =

na + mb where m and n are any pair of chosen numbers such that (m, n) = 1 (m

≠ 1,

≠ 1). Clearly (Z

,.) is a groupoid which is not a semigroup but Z

under usual

multiplication and addition are semigroups .

D

EFINITION

1.3.3: Let L be a non-empty set, we say L is a group if on L is defined a

binary operation ‘

∗’ such that

1. L is closed under ‘

∗’ i.e. a ∗ b ∈ L for all a, b ∈ L.

2. a

∗ (b ∗ c) = (a ∗ b) ∗ c for all a, b, c ∈ L.

3. There exist an element e

∈ L such that a ∗ e = e ∗ a = a for all a ∈ L.

4. For every a

∈ L there exist a unique element a

-1

∈ L such that a ∗ a

-1

= a

-1

∗ a

= e.

A non-associative group is called a loop.

D

EFINITION

1.3.4: The Smarandache semigroup (S-semigroup) is defined to be a

semigroup A such that a proper subset of A is a group.

D

EFINITION

1.3.5: A Smarandache groupoid G (S-groupoid) is a groupoid, which

has a proper subset S

⊂ G which is a semigroup under the operations of G.

EFINITION

1.3.6: A loop L is said to be a Smarandache loop (S-loop) if L has a

proper subset, which is a group under the operations of L.

D

EFINTION

1.3.7: Let R be a non empty set on which is defined two binary

operations '+' and '.' ; (R, +, .) is a non-associative ring, if

1. (R, +) is an additive abelian group with 0
2. (R, .) is closed with respect to '.' and a . (b.c)

≠ (a.b) . c for some a, b, c ∈ R.

3. a. (b + c) = a.b + a.c (The distributive law are true) for all a, b, c,

∈ R.

Let R be a non-assoiciative ring. R is said to be a Smarandache non-associative ring
(SNA-ring) if S contains a proper subset P such that P is an associative ring under
the operations of R.

(We as a matter of convention put ab for a.b).

Loop rings are constructed analogous to group rings. Group rings are associative
where as loop rings are non-associative rings. Similarly analogous to semigroup
rings, groupoid rings are constructed replacing a semigroup by a groupoid. Groupoid
rings are also non-associative rings. Thus the study of there two classes of non-
associative rings is carried out in chapters 2 and 3 respectively.

Chapter 2

LOOP RINGS AND SMARANDACHE LOOP
RINGS

This chapter introduces the notion of Smarandache loop rings. It has five sections. In
section one we just recall the definition of loops and Smarandache loops and give
some of its basic properties which are very essential for the study of later sections in
this chapter.

In section two we recall the properties of loop rings and introduce the basic
definitions and notions of Smarandache loop rings. In fact several properties are
defined and illustrated with examples. Section three defines several types of
Smarandache elements. Around twenty theorems are proved about these special
Smarandache elements. Smarandache substructures are analysed and defined in
section four. Over twenty definitions and thirty-five theorems about these
Smarandache substructures are given. In the final section we give a few general
properties of loop rings and their Smarandache analogues.

2.1 Introduction of Loops and Smarandache loops

In this section we just recall the definition of loops and Smarandache loops, illustrate
them with examples and enumerate its basic properties.

D

EFINITION

2.1.2: A non-empty set L is a said to form a loop, if in L is defined a

binary operation called product and denoted by ‘.’ such that for all a, b

∈ L we have

a.b

∈ L, L contains an element e such that a.e = e.a = a for all a ∈ L; e is called the

identity element of L. For every ordered pair (a, b)

∈ L × L there exists a unique pair

(x, y)

∈ L × L such that ax = b and ya = b. The binary operation ‘.’ need not in

general be associative for a loop.

Thus all groups are loops and loops in general are not groups.

Example 2.1.1: Let (L, .) be a loop given by the following table:

. e a

e e a

e a

EFINITION

2.1.2: Let (L, .) be a loop, we say L is a commutative loop if a.b = b.a

for all a, b

∈ L. Loop L is said to be of finite order if L has finite number of elements;

and we denote it by o(L) or |L|. If L has infinite number of elements then we say L is
of infinite order.

D

EFINITION

2.1.3: Let (L, .) be a loop and S a proper subset of L. We say (S, .) is a

subloop if S itself is a loop under ‘.’ A subloop S of L is said to be a normal subloop
of L if

1.  xS = Sx.
2.  (Hx) y = H (xy).
3.  y (xH) = (yx) H for all x, y

∈ L.

A loop L is said to be a simple loop if it does not contain any nontrivial normal
subloops.

D

EFINITION

2.1.4: If x and y are elements of a loop L, the commutator (x, y) is

defined by xy = (yx) (x, y).

The commutator subloop of a loop L (denoted by L’) is the subloop generated by all
of its commutators, that is L' =

〈{x ∈ L / x = (y, z) for some y, z ∈ L}〉, where L' ⊆ L,

〈L'〉 denotes the subloop generated by L'.

D

EFINITION

2.1.5: Let L be a loop, if x, y, z are elements of a loop L an associator(x,

y, z) is defined by (xy)z = (x(yz)), (x, y, z) the associator subloop of a loop L (denoted
by A (L)) is the subloop generated by all of its associators, that is,

〈 {x ∈ L / x = (a,

b, c) for some a, b, c

∈ L}〉.

EFINITION

2.1.6: A loop L is said to be a Moufang loop if it satisfies any one of the

following identities

(xy)(zx) = (x (yz)) x.
((xy) z) = x (y (xy)).
x (y (xz)) = ((xy) x) z

for all x, y, z

∈ L.

These identities will be termed as Moufang identities.

D

EFINITION

2.1.7: A loop L is called a Bruck loop if (x (yx)) z = x (y (xz)) and (xy)

-1

= x

-1

for all x, y, z

∈ L.

EFINITION

2.1.8: A loop (L, .) is called a Bol loop if ((xy) z) y = x ((yz) y) for all x,

y, z

∈ L.

EFINITION

2.1.9: A loop L is said to be right alternative if (xy) y = x (yy) for all x, y

∈ L and left alternative if (xx)y = x(xy) for all x, y ∈ L. L is said to be an alternative
loop, if it is both a right and a left alternative loop.

D

EFINITION

2.1.10: A loop (L, .) is called a weak inverse property loop (WIP-loop)

if (xy)z = e imply x(yz) = e for all x, y, z

∈ L.

EFINITION

2.1.11: A loop L is said to be semi alternative if (x, y, z) = (y, z, x) for

all x, y, z

∈ L, where (x, y, z) denotes the associator of elements x, y, z ∈ L.

EFINITION

2.1.12: Let L be a loop. The left nucleus N

= {a

∈ L / (a, x, y) = e for

all x, y

∈ L} is a subloop of L. The middle nucleus N

= {a

∈ L / (x, a, y) = e for all

x, y

∈ L} is a subloop of L. The right nucleus N

= {a

∈ L / (x, y, a) = e for all x, y ∈

L} is a subloop of L where e is the identity elemtent of L.

The nucleus N(L) of a loop L is the subloop given by N(L) = N

∩ N

EFINITION

2.1.13: The Moufang center C(L) is the set of elements of the loop L

which commute with every element of L, that is C (L) = {x

∈ L / xy = yx for all y ∈

L}.

The center Z(L) of a loop L is the intersection of the nucleus and the Moufang center
that is Z(L) = C(L)

∩ N(L).

EFINITION

2.1.14: A map

θ from a loop L to a loop L’ is called a homomorphism if

θ (ab) = θ (a) θ (b) for all a, b ∈ L.

D

EFINITION

2.1.15: Let (L, .) be a loop. For

α ∈ L define a right multiplication R

the permutation of the loop (L, .) as follows. R

: x

→ x.a we will call the set {R

/ a

∈

L} the right regular representation of (L, .) or briefly, the representation of L.

D

EFINITION

2.1.16: For any predetermined a, b in the loop L, a principal isotope (L,

∗) of the loop (L, .) is defined by x ∗ y = X.Y where X.a = x and b.Y = y. Let L be a
loop. L is said to be G-loop if it is isomorphic to all of its principal isotopes.

Now we define a new class of loops of even order built using Z

EFINITION

2.1.17: Let L

(m) = {e, 1, 2, 3, …, n} be the set where n > 3, n is odd

and m is a positive integer such that (m, n) = 1 and (m – 1, n) = 1 with m < n. Define
on L

(m) a binary operation ‘.’ as follows.

e.i = i.e = i for all i

∈ L

(m)

i.i = i

= e for all i

∈ L

(m)

i.j = t where t = (mj – (m – 1) i) (mod n)

for all i, j

∈ L

(m); i

≠ j, i ≠ e and j ≠ e.

Then (L

(m), .) is a loop.

Let L

denote the class of all loops L

(m) for a fixed n and various m’s satisfying the

conditions; m < n; (m, n) = 1 and (m – 1, n) = 1; that is L

= {L

(m) / n > 3, n odd,

m < n, (m, n) = 1 and (m – 1, n) = 1}.

Example 2.1.2: Let L

(3) = {e, 1, 2, 3, 4, 5, 6, 7} be a loop in L

given by the

following table:

. e 1 2 3 4 5 6 7
e e 1 2 3 4 5 6 7
1 1 e 4 7 3 6 2 5
2 2 6 e 5 1 4 7 3
3 3 4 7 e 6 2 5 1
4 4 2 5 1 e 7 3 6
5 5 7 3 6 2 e 1 4
6 6 5 1 4 7 3 e 2
7 7 3 6 2 5 1 4 e

Clearly L

(3) is a WIP-loop and it is non-commutative and of order 8.

Now we proceed on to define the concept of Smarandache loops and give some of
the Smarandache properties enjoyed by them as that alone will make this chapter
self-contained.

D

EFINITION

2.1.18: A Smarandache loop (S-loop) is defined to be a loop such that a

proper subset A of L is a subgroup (with respect to the same induced operations) that
is

φ ≠ A ⊂ L.

Example 2.1.3: Let L be a loop given by the following table:

. e 1 2 3 4 5

e e 1 2 3 4 5

1 1 e 4 2 5 3
2 2 4 e 5 3 1
3 3 2 5 e 1 4
4 4 5 3 1 e 2
5 5 3 1 4 2 e

Clearly {e, 1}, {e, 2}, {e, 3}, {e, 4} and {e, 5} are subgroups of L. Hence L

(3) is a

S-loop.

T

HEOREM

[72]: Every power associative loop is a S-loop.

Proof: Obvious by the very definition of power associative loop as every element
generates a group. Hence every power associative loop is a S-loop.

T

HEOREM

[72]: Every diassociative loop is a S-loop.

Proof: Follows from the fact that every pair of elements in a loop generates a
subgroup. Hence the claim.

T

HEOREM

[72]: Every loop L

(m) in L

, n > 3, n an odd integer (m, n) = 1 and (n, m

– 1) = 1 with m < n is a S-loop.

Proof: Since L

(m)

∈ L

are power associative loops as every element generates a

group, we see each L

(m)

∈ L

is a S-loop.

EFINITION

[72]: Smarandache Bol loop (S-Bol loop) is defined to be a loop L such

that a proper subloop (which is not a subgroup) A of L is a Bol loop (with respect to
the operations of L). That is 0

≠ A ⊂ S. Similarly we define Smarandache Bruck loop

(S-Bruck loop), Smarandache Moufang loop (S-Moufang loop) and Smarandache
right (left) alternative loop (S-right (left) alternative loop).

In the definition we insists that A should be a subloop of L and not a subgroup of L.
For every subgroup is a subloop but a subloop in general is not a subgroup. Further
every subgroup will certainly be a Moufang loop, Bol loop, or a Bruck loop.

T

HEOREM

[72]: Every Bol loop is a S-Bol loop but every S-Bol loop is not a Bol

loop.

Proof: Left for the reader to prove.

D

EFINITION

2.1.19: Let L and L' be two S-loops, with A and A' its subgroups

respectively. A map

φ from L to L' is called a Smarandache loop homomorphism (S-

loop homomorphism) if

φ restricted to A is mapped onto the subgroup A' in L', that is

φ: A → A' is a group homomorphism.

The concept of S-loop isomorphism and automorphism are defined in a similar way.
For results about S-loops refer [72]. We conclude this section with the introduction
of Smarandache subloop, Smarandache normal subloop and related concepts.

D

EFINITION

2.1.20: Let L be a loop a proper subset A of L is said to be a

Smarandache subloop (S-subloop) of L if A is a subloop of L and A is itself a S-loop;
that is A contains a proper subset B which is a group under the operations of L, i.e.,
we demand a S-subloop which is not a subgroup.

T

HEOREM

2.1.1: Let L be a loop, if L has a S-subloop then L is a S-loop.

Proof: Straightforward by the definition.

T

HEOREM

2.1.2: Let L be a S-loop every subloop of L need not be a S-subloop of L.

Proof: By an example. Let L

(2)

∈ L

be a loop. It is clearly a S-loop but L

(2) has

no S-subloops.

D

EFINITIONS

2.1.21: Let L be a S-loop. If L has no subloops but only subgroups we

call L a Smarandache subgroup loop (S-subgroup loop).

T

HEOREM

2.1.3: Let L

(m)

∈ L

where n is a prime. Then the class of loops in L

are

S-subgroup loop.

Proof: Use simple number theoretic argument and use the fact n is a prime we see
L

(m) has only subgroups and no subloops.

EFINITION

2.1.22: Let L be a loop. We say a nonempty subset A of L is a

Smarandache normal subloop (S-normal subloop) if

a. A is itself a normal subloop of L.

b. A contains a proper subset B where B is a subgroup under the operations of

L. If L has no S-normal subloop we say the loop L is Smarandache simple (S-
simple)

HEOREM

2.1.4: Let L be a loop. If L has a S-normal subloop then L is a S-loop.

Proof: Left for the reader as an exercise.

D

EFINITION

2.1.23: Let L be a finite S-loop. An element a

∈ A, A ⊂ L, A the

subgroup of L is said to be Smarandache Cauchy element (S-Cauchy element) of L if
a

= 1 (r > 1), 1 is the identity element of L and r divides the order of L; otherwise a

is not a S-Cauchy element of L.

D

EFINITION

2.1.24: Let L be a finite S-loop. If every element in every subgroup is a

S-Cauchy element then we say S is a Smarandache Cauchy loop (S-Cauchy loop).

T

HEOREM

[72]: Every loop in the class of loops L

are S-Cauchy loops.

Proof: Obvious by the very definitions.

D

EFINITION

2.1.25: Let L be a finite loop. If the order of every subgroup in L divides

the order of the loop L then L is a Smarandache Lagranges loop (S-Lagranges loop).

D

EFINITION

2.1.26: Let L be a finite loop. If there exist at least one subgroup in L

whose order divides the order of L then L is Smarandache weakly Lagranges loop
(S- weakly Lagranges loop).

D

EFINITION

2.1.27: Let L be a loop. We say L is a Smarandache commutative loop

(S-commutative loop) if L has a proper subset A such that A is a commutative group.

Example 2.1.4: The loop L

(3)

∈ L

is a S-commutative loop but clearly L

(3) is not

a commutative loop.

D

EFINITION

2.1.28: If every subgroup of a loop L is commutative we call L a

Smarandache strongly commutative loop (S-strongly commutative loop).

The author is requested to obtain the relations between these concepts.

T

HEOREM

[72]: Let L be a power associative loop, L is then a S-commutative loop.

Proof: Direct by the very definitions.

Thus we have a class of loops which are S-commutative.

D

EFINITION

2.1.29: Let L be a loop. L is a Smarandache cyclic loop (S-cyclic loop)

if L contains atleast a proper subset A which is a cyclic group. We say L is

Smarandache strongly cyclic (S-strongly cyclic) if every subset which is a subgroup
is a cyclic group.

T

HEOREM

[72]: Let n be a prime, every loop in the class of loops L

are S-strongly

cyclic loops.

Proof: Follows from the fact when n is a prime, L

(m) for any m have only n

subgroups each cyclic of order 2.

T

HEOREM

[72]: Let L

be a class of loops, n > 3, if

...

(

≥ 1 for i =

1, 2, 3, …, k) then it contains exactly F

loops which are strictly non-commutative

and they are

S-strongly commutative loops

S-strongly cyclic loops; where

)

(

−

Proof: Kindly refer [72].

D

EFINITION

2.1.30: Let L be a loop. If A (A a proper subset of L) is a S-subloop of L

is such that A is a pseudo commutative loop then we say L is a Smarandache pseudo
commutative loop (S-pseudo commutative loop) i.e., for every a, b

∈ A, we have x ∈

B such that a(xb) = b (xa) (or (bx) a) B a subgroup in A.

If a.x.b = b.x.a for all x

∈ A and a, b ∈ A we say L is a Smarandache strongly

pseudo commutative loop (S-strongly pseudo commutative loop). Now we proceed on
to define the concept of Smarandache commutator subloop (S-commutator subloop)
of a loop L denoted by L

. Let L be a loop; A

⊂ L be a S-subloop of L, L

〈 {x ∈ A /

x = (y, x) for some y, z

∈ A}〉 with the usual notation and if L has no S-subloops but L

is a S-loop we replace A by L.

T

HEOREM

[72]: Let L be a S-loop, which has no S-subloops then L

= L

Proof: Left for the reader to prove.

Similarly we have to define the concept of Smarandache associative subloop of a
loop L.

D

EFINITION

2.1.31: Let L be a loop. We say L is a Smarandache associative loop (S-

associative loop) if L has a S-subloop A such that, A contains a triple x, y, z (all three
elements distinct and different from e, the identity element of A) such that x.(y.z) =
(x.y).z. This triple is called the Smarandache associative triple (S-associative triple).

We say L is a Smarandache strongly associative loop (S-strongly associative loop) if
in L every S-subloop has an S-associative triple.

Suppose if x, y

∈ A (A ⊂ L, L a loop, A a S-subgroup of L) we have (xy) x = x (yx)

then we say the loop L is a Smarandache pair wise associative loop (S-pairwise
associative loop).

The Smarandache associator subloop (S-associator subloop) of L is denoted by L

, is

the subloop generated by all the associators in A, where A is a S-subloop of L (A

⊂

L) i.e., L

〈{x ∈ A / x = (a, b, c) for some a, b, c ∈ L}〉. If L has no S-subgroup but

L is a S-loop we replace A by L itself. Thus we have in that case the S-associator
coincides with the associator of the loop L.

D

EFINITION

2.1.32: Let L be a loop, A is a S-subloop in L, if we have an associative

triple a, b, c in A such that (ax) (bc) = (ab) (xc) for some x

∈ A, we say L is a

Smarandache pseudo associative loop (S-pseudo associative loop) if every
associative triple in A is a pseudo associative triple of A for some x in A. If L is a S-
loop having no S-subloops then we replace A by L. Suppose L is a S-pseudo
associative loop, then PA (L

) =

〈{t ∈ A / (ab) (tc) = (at) (bc) where (ab) c = a (bc)

for a, b, c

∈ A}〉 denotes the Smarandache pseudo associator (S-pseudo associator).

We define or recall the definition of these concepts as they will play a major role in
the definition and study of properties of the Smarandache non-associative rings,
which is very recent [73].

P

ROBLEMS

1.  Give an example of a S-loop of order 15.
2.  Can a S-loop of order 5 be S-cyclic?
3.  Given an example of a S-loop of order 21, which is S-commutative.
4. Find L

for the loop L = L

(5).

5. Does L

(5) have nontrivial L

2.2 Properties of Loop Rings and Introduction to Smarandache loop
rings

In this section we just recall the definition of loop rings and enumerate several of the
properties enjoyed by loop rings and here give some of its basic Smarandache
properties mainly using the well known class of loops L

. Further we state that

throughout this paper by a ring R we mean only a commutative ring with 1 or a field;
it can be finite or infinite. Z

denotes the ring of integers modulo n, n a composite

number, Z

, the prime field of characteristic p, Z the set of integers which is a ring, in

fact an integral domain, Q the field of rationals and R the field of reals. L will denote
a loop non-necessarily abelian but never a group so by a loop we mean a non-
associative group.

D

EFINITION

2.2.1: Let R be a commutative ring with identity 1 and let L be a loop.

The loop ring of the loop L over the ring R denoted by RL consists of all finite formal
sums of the form

∑

(i – runs over a finite number) where

∈ R and m

∈ L

satisfying the following conditions.

∑

⇔

for i = 1, 2, 3, …, n

∑

























)

(

























∑

where

Σ α

= m

for all r

∈ R and m

∈ L.

∑

)

(

for r

∈ R and Σr

∈ RL.

RL is a non-associative ring with 0

∈ R as its additive identity. Since 1 ∈ R we have

L = 1.L

⊆ RL and Re = R ⊆ RL, where e is the identity of L.

We give one example of a loop ring.

Example 2.2.1: Let Z

= {0, 1} be the prime field of characteristic two and L be the

loop given by the following table:

. e a b c d
e e a b c d
a a e c d b
b b d a e c
c c b d a e
d d c e b a

L is the loop ring, which is obviously a non-associative ring.

We give some of the basic definitions about loop rings.

D

EFINITION

2.2.2: A loop ring RL is said to be an alternative loop ring if (xx) y =

x(xy) and x (yy) = (xy) y for all x, y

∈ RL. In a ring R let x, y ∈ R. A binary operation

known as the circle operation is denoted by ‘o’ and is defined by x o y = x + y – xy.

D

EFINITION

2.2.3: Let x

∈ R, R a ring. x is said to be right quasi regular (r.q.r) if

there exist a y

∈ R such that x o y = 0 and x is said to be left quasi regular (l.q.r) if

there exists a y'

∈ R such that y' o x = 0. An element is quasi regular (q.r) if it is both

right and left quasi regular. y is known as the right quasi inverse (r.q.i) of x and y' is
the left quasi inverse (l.q.r) of x. A right ideal or left ideal in R is said to be right
quasi regular (l.q.r or q.r respectively) if each of its element is right quasi regular
(l.q.r or q.r respectively).

D

EFINITION

2.2.4: Let R be a ring. An element x

∈ R is said to be a regular element

if there exists a y

∈ R such that xyx = x.

Since loop rings are non-associative we need to define left regular element and right
regular elements for loop rings.

D

EFINITION

2.2.5: Let RL be a loop ring of a finite loop L over the ring R. An

element x

∈ RL is said to be right regular (left regular) if there exists a y ∈ RL (y' ∈

RL) such that x (yx) = x or ((xy') x = x).

The Jacobson radical J(R) of a ring R is defined as follows:

J(R) = {a

∈ R / a R is right quasi regular ideal of R}.

A ring R is said to be semi-simple if J(R) = {0}, where J(R) is the Jacobson radical
of R.

D

EFINITION

2.2.6: Let RL be a loop ring of a loop L over the ring R. Let

∑

∈

Support of

α denoted by supp α = {m

≠ 0}. |Supp α| denotes the number of

elements in support of

α. The augmentation ideal of RL denoted by W(RL) = {α =

Σα

∈ RL / Σα

= 0}.

Now we recall some of the basic properties starting from some of the properties
enjoyed by the new class of loops over rings Z

or Z

or Z or Q.

HEOREM

2.2.1: Let L be a loop of order n and R be any commutative ring with 1

and of characteristic 0. RL be the loop ring of the loop L over R. If

α = Σα

is right

quasi regular, then

Σ α

≠ 1 (α

∈ R and m

∈ L for all i = 1, 2, 3, …, n).

Proof: It is a matter of routine once we take L = {m

= 1, m

, …, m

} to be the n

elements of L and make use of the fact, for

∑

in RL;

α + β – αβ = 0. Equating the coefficients of the like terms and adding these

coefficients we get

Σα

Σβ

– (

Σα

)(

Σ β

) = 0 or

Σ α

Σβ

(

Σα

– 1). If

Σα

= 1 then

we get

Σ α

= 0 a contradiction hence

Σ α

≠ 1. When we replace the ring of integers

R by Z we get if

α = Σα

∈ ZL is right quasi regular only in two cases when

Σ α

= 0 or

Σ α

= 2. A similar argument will yield the result directly.

Finally in case of finite characteristic we arrive at a result only in case of the prime
field of characteristic p; i.e., Z

. Let Z

L be the loop ring,

Σα

is right quasi regular

only when

Σ α

≡/ 1 (mod p).

HEOREM

2.2.2: Let L be a loop and Z

L be the loop ring. (Z

= (0, 1) the prime

field of characteristic two). If

α ∈ Z

L is right quasi regular, then |supp

α| is an even

number.

Proof: Left for the reader as an exercise.

T

HEOREM

2.2.3: Let Z

be the prime field of characteristic two. L a commutative

loop in which every element m

∈ L is such that m

= 1, then

α = Σ α

∈ Z

L is

right quasi regular if and only if |supp

α| is an even number.

Proof: Direct by using the fact m

= 1 for every m

∈ L and L is a commutative loop.

Thus we have still modified results in case of field of prime characteristic p, p an odd
prime.

Example 2.2.2: Let L be a loop given by the following table:

. e a b c d

e e a b c d
a a e c d b

b b d a e c

c c b d a e

d d c e b a

Z the ring of integers. ZL be the loop ring of a loop L over Z. e + a

∈ ZL is quasi

regular for (e + a) o (e + a) = (e + a) + (e + a) – (e + a) (e + a) = 0. Consider x = a – d
∈ ZL with y = e + a – c + d we get (a – d) o (e + a – c + d) = a – d + (e + a – c + d) –
(a – d) (e + a – c + d) = 0 and

(e + a – c + d) o (a – d) = e + a – c + d + a – d – (e + a – c + d) (a – d) = 0. Thus a – d
is also quasi regular, now consider

(e + a) 0 (a – d) = e + a + a – d – (e + a) (a – d) = a + b.

It is easily verified a + b has no right quasi inverse and left quasi inverse in ZL. Thus
we make the following important observation that in general the set of quasi regular
elements of a ring need not form a closed set under the circle operation.

One of the nice results about Jacobson radical for loops rings is

Result: Let L be a finite loop. Z

the prime field of characteristic 2 and Z

L be the

loop ring of the loop L over Z

. Then m

+ m

∉ J(Z

L) (where |L| = n, n is odd, n

≥

7; Z

= [0, 1]) for any m

, m

∈ L.

HEOREM

2.2.4: Let F be a field of characteristic zero and L a finite loop; FL the

corresponding loop ring. If

α = Σ α

∈ FL (Σα

≠ 0) is a right (left) regular

element with y as its right (left) relative inverse, then

Σ β

= 1/

Σ α

Proof: Let

α be the right regular with y as its right relative inverse. Then x (yx) = x

that is

Σ α

((

Σ β

) (

Σα

)) =

Σ α

. Equating the coefficients of the like

terms and adding all the equations we get (

Σα

) ((

Σβ

) (

Σα

)) =

Σ α

Σβ

= 1 /

Σα

(as

Σα

≠ 0).

HEOREM

2.2.5: Let Z

L be the loop ring of a finite loop L = {m

= e, m

, …, m

}

over the prime field Z

. Then x = k (e + m

+ … + m

) (0

≠ k ∈ Z

) is regular if and

only if p / n where n is the order of the loop.

Proof: Left for the reader as an exercise.

In view of this it can be easily proved that if p/n then Z

L is not regular.

HEOREM

2.2.6: Let R be the field of reals and L a finite loop in which m

= 1 for

every m

∈ L. Then the loop ring RL has a nontrivial idempotent α = Σ α

only if

< 1.

Proof: By simple number theoretic arguments we get the result as

+ … +

< 1.

HEOREM

2.2.7: Let Z

L be the loop ring of a loop L over Z

. Then x = am

+ bm

∈

L (a

≠ 0, a ≠ b, b ≠ 0, m

≠ m

with p an odd prime) is an idempotent if and only if

= e and m

= e, a = (p + 1) / 2 and b = (p – 1) / 2.

Proof: Simple calculations using number theoretical concepts, the proof is easily
obtained. Hence left for the reader to prove.

We now give some relation between the quasi regular elements and idempotents,
how the existence of one, results in the non-existence of the other which is given by
the following theorem. The reader is expected to prove it.

T

HEOREM

2.2.8: Let F be a field of characteristic two and let L = {m

= e, m

, …,

} be the finite loop of order n. FL the loop ring of L over F. Then

α = t (e + m

… + m

) (t

≠ 0) is a quasi regular element in FL if and only if α is not an idempotent

element in L.

Yet another result of some significance is.

T

HEOREM

2.2.9: FL be the loop ring of a finite loop L in which there exists at least

one m

in L such that m

= e (m

≠ e) and F be a field of characteristic zero. Then α

= m (e + m

)

∈ FL (m ≠ 0) is quasi regular if and only if α is not an idempotent.

Proof: Straightforward, hence left for the reader to prove.

Thus we can conclude if L is a loop such that for each m

∈ L we have m

= 1 and Z

be the ring of integers then in the loop ring ZL no non-zero idempotent can be right
quasi regular. Further even in Z

L, L satisfying the condition m

= 1 for all m

in L

and |L| <

∞ a non-zero idempotent element cannot be right quasi regular.

Now we give the relation between the J(FL) and W(FL) in a loop ring.

T

HEOREM

2.2.10: Let L be a finite loop and let F be any field of characteristic zero

or ring of integers. Then J(FL)

⊂ W(FL).

Proof: The reader is expected to prove.

Now a nontrivial question would be; can we ever have J(FL) = W(FL)? The answer
is yes; given by the following theorem, which the reader is expected to prove.

T

HEOREM

2.2.11: Let F be any field or ring of integers and L be a finite loop. Then

W(FL) = J(FL) if and only if every element of W(FL) is right quasi regular.

This is however illustrated by the following example.

Example 2.2.3: Let Z

= {0, 1} be the prime field of characteristic two and L be a

loop given by the following table:

. e x

e e x

e x

It is easily verified J(Z

L) = W(Z

L). If we replace Z

by Z

we see that J(Z

⊂

W(Z

L). Z be the ring of integers and let L be a finite loop, then J(ZL)

⊂ W(ZL).

Several such results can be obtained. Now we proceed on to define Smarandache
loop ring we in fact will define two levels of Smarandache loop rings.

D

EFINITION

2.2.7: Let L be a loop and R any ring. The loop ring RL is a

Smarandache loop ring (S-loop ring) if and only if L is a S-loop.

Thus in the very definition itself we demand the loop to be a S-loop. Even if the loop
ring RL is a SNA-ring still we will not accept the ring to be a S-loop ring.

A natural question would be what is a SNA-ring. We first define it in the following.
We first define level I Smarandache NA rings.

D

EFINITION

2.2.8: Let (R, +, .) be a non-associative (NA) ring. If a proper subset P

of R is an associative ring under the operation of ‘+’ and ‘.’ then we say R is a
Smarandache NA ring (SNA-ring) of level I. That is if a ring is properly contained in
a non-associative ring then we call the NA ring to be a Smarandache NA-ring (SNA-
ring).

Example 2.2.4: Let L be a loop given by the following table:

. e a

e e a

e a

Z be the ring of integers, ZL be the loop ring, clearly ZL is a non-associative ring.
Now 1.L = L

⊂ ZL; so ZL is a SNA-ring. A natural question is "Will every loop ring

be a SNA-ring?" The answer is yes as every loop contains 1 and the ring R over
which the loop ring is defined is an associative ring. Hence every loop ring is a SNA-
ring of level II.

But is the loop ring a S-loop ring the answer is need not be. For if the loop L fails to
be a S-loop, the loop ring will not be a S-loop ring even if it happens to be a SNA-
ring.

D

EFINITION

2.2.9: Let R be a NA-ring. A non-empty subset S of R is said to be a

SNA-subring of R if S contains a proper subset P such that P is an associative ring
under the operations of R.

T

HEOREM

[73]: Let R be a NA-ring. If R has a SNA-subring then R is a SNA-

subring.

Proof: Direct by the very definitions.

T

HEOREM

2.2.12: Let (R, +, .) be a SNA-ring. Every subring of R need not in

general be a SNA-subring of R.

Proof: Let ZL be the loop ring given in example 2.2.4. Clearly if we take H = {e, a

}

then ZH is a subring of ZL which is not a SNA-subring of ZL. Hence the claim.

Now we proceed on to define the concept of Smarandache ideal in a SNA-ring.

D

EFINITION

2.2.10: Let R be a non-associative ring. A proper subset I of R is said to

be a Smarandache NA right ideal (SNA-right ideal) of R if

I is a SNA-subring of R; i.e., J

⊂ I, J is a proper subset of I which is an associative

subring under the operations of R.

For all i

∈ I and j ∈ J we have either ij or ji is in J. If I is simultaneously a SNA-

right ideal and SNA-left ideal then we say I is a SNA-ideal of R.

HEOREM

2.2.13: Let R be a non-associative ring. If R has a SNA-ideal then R is a

SNA-ring.

Proof: Straightforward by the very definition of SNA-ideal. Another interesting
result is that in general a SNA-ideal need not be an ideal.

In view of this we have the following.

T

HEOREM

2.2.14: Let R be any non-associative ring. I be a SNA-ideal of R then I in

general need not be an ideal of R.

Proof: By an example, we can prove; the task of which is left for the reader.

D

EFINITION

2.2.11: Let R be a NA ring we say R is a Smarandache NA Moufang

ring (SNA-Moufang ring) if R contains a subring S where S is a SNA-subring and for
all x, y, z in S we have (x.y).(z.x) = (x.(y.z)).x that is the Moufang identity is true in S.

It is to be noted that we do not demand the whole of R to satisfy Moufang identity. It
is sufficient if a S-subring of R satisfies the identity, that is enough for the ring to be
a SNA-Moufang ring.

D

EFINITION

2.2.12: Let R be a non-associative ring, R is said to be a Bol ring if R

satisfies the Bol identity ((x

∗y) ∗ z) ∗ y = x ∗ ((y ∗ z) ∗ y) for all x, y, z ∈ R.

We now define the condition for it to be a SNA-Bol ring.

D

EFINITION

2.2.13: Let R be a non-associative ring, R is said to be SNA-Bol ring if R

contains a subring S such that S is a SNA-subring of R and we have the Bol identity
to be true for all x, y, z

∈ S ⊂ R.

In view of this we have the following theorem:

T

HEOREM

2.2.15: Let R be a non-associative ring which is a Bol ring. Then R is a

SNA-Bol ring.

Proof: Direct by the very definitions.

T

HEOREM

2.2.16: Let R be a non-associative ring which is a SNA-Bol ring then R

need not be a Bol ring.

Proof: The reader is requested to construct an example to prove this theorem.

D

EFINITION

2.2.14: Let R be any non-associative ring. R is said to be a right

alternative ring if (xy) y = x (yy) for all x, y

∈ R similarly left alternative if (xx) y = x

(xy) for all x, y

∈ R. Finally R is an alternative ring if it is simultaneously both right

alternative and left alternative.

Example 2.2.5: Let Z be the ring of integers. L be a loop given by the following
table:

. e g

e e g

e g

Clearly it can be easily verified that the loop ring ZL is a right alternative ring.

D

EFINITION

2.2.15: Let R be a ring; R is said to be a SNA-right alternative ring if R

has a subring S such that S is a SNA-subring of R and S is a right alternative ring
that is (xy) y = x (yy) is true for all x, y

∈ S. Similarly we can define SNA-left

alternative ring. If R is simultaneously both SNA-right alternative and SNA-left
alternative ring then we say R is a SNA-alternative ring.

D

EFINITION

2.2.16: Let R be a non-associative ring. R is said to be a SNA-

commutative ring if R has a proper subset S (S

⊂ R) such that S is a commutative

associative ring with respect to the operations of R.

D

EFINITION

2.2.17:Let (R, +, .) and (R

, +, .) be any two SNA-rings. A map

φ : R to

is a Smarandache NA-ring homomorphism (SNA-ring homomorphism) if

φ : A →

is a ring homomorphism where A

⊂ R and A

⊂ R

are just associative subrings of

R and R

, respectively. Clearly Ker

φ is an ideal of A and never an S-ideal of R. In

fact Ker

φ may not be even an ideal of R.

EFINITION

2.2.18: Let R be a SNA-ring if R has no SNA-ideals then we say R is

SNA-simple ring.

We define SNA-quotient ring for a given ring R and an SNA-ideal I; R / I is a
Quotient ring as in the case of rings. One is not in a position to say whether the SNA-
quotient rings of a SNA-maximal ideal is simple? The answer to this question is little
complicated as we have the very definition of SNA-ideals only relative to SNA-
subrings and not relative to the whole ring R. So one may presume that the quotient
ring may have SNA-ideals relative to some other SNA-subrings.

Thus when we say the ring R is SNA-simple it is assumed the ring has no SNA-
ideals related to any of the SNA-subrings. One may be wanting to know the concept
of Smarandache maximal and SNA-minimal ideals. For the sake of helping the
reader we define explicitly these types of ideals.

D

EFINITION

2.2.19: Let R be a ring. S a SNA-subring of R. We say an SNA-ideal I

⊂

S is maximal if we have an SNA-ideal, K relative to S such that I

⊂ K ⊂ S then either

I = K of K = S similarly an ideal J

⊂ S is minimal if we have another ideal L ⊂ S

such that L

⊂ J then L = (0) or L = J.

Unlike in the case of minimal and maximal ideals we see in case of SNA-ideals the
maximality or the minimality is only a relative concept for it follows by the very
definition of them. Similarly we say a SNA-ideal is principal if it is generated by a
single element. We say a SNA-ideal J of S, S a SNA-subring of R is prime if xy

∈ J

then x or y is in J.

P

ROBLEMS

1. Let L = {m

= e, m

, …, m

} be a finite loop with m

= e for all m

∈ L. Z

be the field of characteristic 2 and Z

L be the loop ring. If

α ∈ Z

L is an

idempotent of Z

L prove

a. |supp

α| is an even number and e ∉ supp α (e the identity element of

L).

b. |supp

α| is an odd number and e ∈ supp α.

c. If L is commutative then Z

L has only trivial idempotents.

2. Let L be a loop given by the following table L

(4).

. e 1 2 3 4 5 6 7
e e 1 2 3 4 5 6 7
1 1 e 5 2 6 3 7 4
2 2 5 e 6 3 7 4 1
3 3 2 6 e 7 4 1 5
4 4 6 3 7 e 1 5 2
5 5 3 7 4 1 e 2 6
6 6 7 4 1 5 2 e 3
7 7 4 1 5 2 6 3 e

Does Z

(4) have idempotents other than 1 and 0? Justify your answer.

3. Can Z

(4) given in problem 2 have nontrivial

a.  quasi regular elements?
b.  regular elements?
c.  right quasi regular elements which are not left quasi regular?

4. Let L

(8) be a loop and Z be the ring of integers. ZL

(8) be the loop ring of

the loop L

(8) over Z.

a. Can

(8) have idempotents? If so find them.

b. Does ZL

(8) have right quasi regular elements?

c. Prove no non-zero idempotent element of ZL

(8) can be right quasi

regular.

5. Let Z

= {0, 1, 2} be the prime field of characteristic 3 and L

(8) be the loop.

(8) be the loop ring of the loop L

(8) over Z

, find

a. W (Z

(8)).

b. J (Z

(8)).

6. For the loop ring Z

L where L is given by the following table:

. e 1 2 3 4 5
e e 1 2 3 4 5
1 1 e 3 5 2 4
2 2 5 e 4 1 3
3 3 4 1 e 5 2
4 4 3 5 2 e 1
5 5 2 4 1 3 e

Find

a. SNA-subrings of Z

b. SNA-ideals if any in Z

c. SNA-principal ideal if any.
d. SNA-maximal / minimal ideal if any.

7. Find for the loop ring ZL where the loop L is given in problem (6) and Z the

ring of integers SNA-subrings which has no SNA-ideals related to them.

8. Does there exists for loop ring Z

(5) and Z

(8) maps

φ such that φ is a

SNA-homomorphism? Find Ker

: Z

→ Z

where A

= {e, 7}

and A

= {e, 2} are subgroups of the loops L

(5) and L

(8) respectively.

Can

be an isomorphism? Justify your claim.

9. Find the ideals of the loop ring Z

(3).

a. Do the two sided ideals of Z

(3) form a modular lattice?

b. Find only the right ideals of Z

(3). What is its lattice structure?

10. For the loops rings Z

(2) and Z

(8) find

a.  The set of right ideals and its lattice structure.
b.  The set of left ideals and their lattice structure.
c.  Two sided ideals and their lattice structure.

11. Find for the loop rings Z

(2) and Z

(8), J (Z

(2)) and J (Z

(8)).

12. Find for the loop rings given in problem 11, W (Z

(2)) and W (Z

(8)).

2.3 Smarandache Elements in Loop Rings

In this section we define the concept of Smarandache zero divisors, Smarandache
nilpotents, Smarandache idempotents, Smarandache units, Smarandache normal

elements, Smarandache semi-idempotents and Smarandache quasi-regular elements
in NA-rings. Since the concept of these Smarandache elements do not in any way
demand the NA-ring to be SNA-ring we proceed on to introduce these notions in any
NA-rings.

D

EFINITION

2.3.1: Let R be a NA-ring with unit. We say x

∈ R \{1} is a Smarandache

unit (S-unit) in R if there exist a y

∈ R with xy = 1.

There exist a, b

∈ R \ {x, y, 1} such that

a.  xa = y or ax = y.
b.  yb = x or by = x and
c.  ab = 1.

HEOREM

2.3.1: Let R be a NA ring. If x is a SNA-unit then x is a unit.

Proof: Direct from the very definition.

It is left for the reader to construct an example of a unit which is not a SNA-unit.

T

HEOREM

2.3.2: Let R be a NA-ring. All units in R in general need not be SNA-units

of R.

Proof: To prove this theorem we construct the following example. Let Z

= {0, 1, 2,

3} be the ring of integers modulo 4. Let L be a loop given by the following table:

. e g

e e g

e g

Let us take the loop ring Z

L identifying 1 with e we assume 1

∈ Z

α = 1 + 2g

∈

L (1 + 2g

)

= 1 is a unit in Z

L but it is not a S-unit of Z

L Thus every unit in

general is not a S-unit.

D

EFINITION

2.3.2: Let R be a non-associative ring. An element a

∈ R \ {0} is called a

Smarandache idempotent (S-idempotent) of R if a

= a. There exists b

∈ R \ {a} such

that

a. b

= a.

b. ab = b or ba = (ba = a) (ab = a).

'or' in the mutually exclusive sense.

HEOREM

2.3.3: Let R be a non-associative ring. If x

∈ R is a S-idempotent then it is

an idempotent of R.

Proof: Straightforward by the very definition.

To find the converse we study and analyze the following example:

Example 2.3.1: Let Z

= {0, 1, 2, 3, 4} be the prime field of characteristic 5. Let

(3) be the loop given by the following table:

. e g

e e g

e g

Consider

α = 1 + g

+ g

in Z

(3). Clearly

α. Thus α is an

idempotent. Take

β = 3 + 3g

= (3 + 3g

)

= 3 + 3g

. (3 + 3g

) (1 + g

+ g

) = 1 + g

+ g

Take X = 4 + 4g

+ 4g

α, further Xα = X. Thus α is a S-idempotent but X is not an idempotent. β = 3

+ 3g

, take Y = 2 + 2g

we have Y

= 3 + 3g

βY = Y.

Thus

β is a S-idempotent. Hence this example shows us the existence of

idempotents, which are S-idempotent in a NA-ring. Now we give a nice theorem for
the existence of S-idempotents in case of loop rings Z

(m) for some suitable n.

HEOREM

2.3.4: Let Z

= {0, 1, 2, .., p-1} be the prime field of characteristic p (p >

2). L

(m) be the loop where we choose n = p. Then

α = (1 + g

+ … + g

) and













;

∈ L

(m) are S-idempotents of the loop ring Z

(m) for any loop L

(m)

∈ L

Proof: Let

α = (1 + g

+ … + g

) be in Z

(m), clearly

α, so α is an idempotent

in Z

(m). Take X = (p – 1) + (p – 1)g

+ … + (p – 1)g

. Clearly X

α and Xα =

X. Hence

α is a S-idempotent take

;

∈ L

(m). Easily checked

β so β is an idempotent in Z

(m). Consider

−

;

βY = Y. Thus elements of the form α and β in Z

(m) are S-idempotents.

EFINITION

2.3.3: Let R be a NA ring. A element a

∈ R \ {0} is said to be a

Smarandache zero divisor (S-zero divisor) if a.b = 0 for some b

≠ 0 in R, and there

exists x, y

∈ R \ {0, a, b} x ≠ y; such that

a. a.x = 0 or x.a = 0.
b. b.y = 0 or y.b = 0.
c. x.y

≠ 0 or y.x ≠ 0.

HEOREM

2.3.5: Let R be a NA-ring. If x

∈ R \ {0} is a S-zero divisor then x is a zero

divisor in R.

Proof: Direct by the very definition of S-zero divisors. Now it may happen that a S-
zero divisor may satisfy all the three conditions a, b, c still it may behave in an
entirely different way.

We define Smarandache pseudo zero divisors.

D

EFINITION

2.3.4: Let R be a NA-ring Let x

∈ R \ {0} be a zero divisor in R. i.e.,

there exists y

∈ R \ {0} with x.y = 0. We say x is a Smarandache pseudo zero divisor

(S-pseudo zero divisor) if their exist a

∈ R \ {x, y, 0} with

a.y = 0 or a.x = 0.
a

= 0.

HEOREM

2.3.6: Let R be a NA-ring. If x

∈ R is a S-pseudo zero divisor of R then x

is a zero divisor of R.

Proof: Follows from the fact that if x is a S-pseudo divisor then it is a zero divisor.

We illustrate them with examples.

Example 2.3.2: Let Z

= {0, 1} be the prime field of characteristic two. L

(3) be the

loop given by the following table:

. e 1 2 3 4 5

e e 1 2 3 4 5

1 1 e 4 2 5 3
2 2 4 e 5 3 1
3 3 2 5 e 1 4
4 4 5 3 1 e 2
5 5 3 1 4 2 e

We shall denote i by g

and e by 1. Let Z

(3) be the loop ring of the loop L

(3)

over Z

. Clearly x = 1 + g

∈ Z

(3) is a zero divisor for y = 1 + g

+ g

is such that xy = 0 and a = 1 + g

is such that a.y = 0 with a

= 0 so x is a S-pseudo

zero divisor and is not a S-zero divisor. It is an obvious zero divisor in Z

(3).

HEOREM

2.3.7: Let R be a NA-ring a S-pseudo zero divisor in general is not a zero

divisor.

Proof: From the above example we see x = 1 + g

is a S-pseudo zero divisor and is

not a S-zero divisor of Z

(3).

In view of this example we have a nice theorem which guarantees the existence of S-
pseudo zero divisors in loop rings using the new class of loops.

T

HEOREM

2.3.8: Let Z

= {0, 1} be the prime field of characteristic two and L

(m)

∈

be a loop in the class of loops. Then all elements of the form x = 1 + g

∈ Z

(m)

where g

∈ L

(m) are S-pseudo zero divisors of Z

(m).

Proof: Given x = 1 + g

∈ Z

(m), clearly for y = 1 + g

+ … + g

we have xy = 0 so

x is a zero divisor in Z

(m). Take a = (1 + g

), i

≠ j with a.y = 0 and we have a

= 0.

Hence x = 1 + g

is a S-pseudo zero divisor of Z

(m).

We still modify by replacing Z

by Z

where m = 2p, p an odd prime; in this case we

have the following results:

T

HEOREM

2.3.9: Let L

(m)

∈ L

be a loop Z

be the ring of integers modulo t where t

= 2p where p is an odd prime. Z

(m) is the loop ring. x = p + pg

∈ Z

(m) is a S-

pseudo zero divisor for all g

∈ L

(m).

Proof: Let x = p + pg

∈ Z

(m), consider y = 1 + g

+ … + g

; we have x.y = 0.

Take a = p + pg

≠ j) then ay = 0 with a

= 0, hence the claim.

The following have been proved about zero divisor in case of power associative and
diassociative loop algebras.

T

HEOREM

[56]: Let K be a field, L a simply ordered loop. Then the loop algebra KL

has no nontrivial zero-divisors.

Proof: Left for the reader to refer [56].

T

HEOREM

[56]: Let K be a field, L a finite power associative or diassociative loop.

Then the loop ring KL has nontrivial divisors of zero.

Proof: KL be the loop algebra. Since L is a diassociative or a power associative loop
we have a proper subgroup H; H

⊂ L. Now KH is a group algebra; since |H| < ∝ we

have for g

∈ H, g

= 1.

Hence (1 – g) (1 + g + … + g

n-1

) = 0. a nontrivial divisor of zero.

In view of these we have the following theorem:

T

HEOREM

2.3.10: Let KL be a S-loop ring where L is a diassociative loop or a

power associative loop of finite order. Then KL has divisors of zero.

Proof: Obvious

Does the loop ring KL have S-divisors of zero?

We define Smarandache weak zero divisor in the following.

D

EFINITION

2.3.5: Let R be a NA ring. An element x

∈ R \ {0} is a Smarandache

weak zero divisor (S-weak zero divisor) if there exist a y

∈ R \ {0, x} such that x.y =

0 satisfying the following condition:

There exists a, b

∈ R \ {x, y, 0} such that

a.x = 0 or x.a = 0
b.y = 0 or y.b 0
x.y = 0 or y.x = 0

Thus in case of S-weak divisor we have another pair of elements which is a zero
divisor.

T

HEOREM

2.3.11: Let K be a field and L a diassociative finite loop or a power

associative finite loop. The loop ring KL has S-weak divisors of zero.

Proof: We have KL to be a S-loop ring. Also g

∈ L is such that g

= 1 and g

∈ H ⊂ L

where H is a group.

(1 – g) (1 + g + … + g

n-1

) = 0

Take a = 1 – g

, r

≠ 0, r > 1

(1 – g

) (1 + g + … + g

n-1

) = 0

Let b = 3 + 3g + … + 3g

n-1

, then b(1 – g) = 0. Now a.b = 0. Hence KL has S-weak

divisors of zero.

T

HEOREM

2.3.12: A S-weak zero divisor of a loop ring in general is not a S-zero

divisor but it is a zero divisor.

Proof: Follows from the definition that S-weak zero divisor is a zero divisor. From
theorem 2.3.11 it is clearly not a S-zero divisor only a S-weak zero divisor.

It is natural to see that if an element is a zero divisor still it need not be S-weak zero
divisor. The reader is advised to construct examples.

EFINITION

2.3.6: Let KL be a loop ring of a loop L over the field K, an element

α ∈

KL is called a normal element of KL if

αKL = KLα. If every element in KL is a

normal element then we say KL is a normal loop ring.

D

EFINITION

2.3.7: Let KL be a loop ring. S be a proper subring of KL, we say S is a

normal subring with respect to a subset T of KL if

xS = Sx for all x

∈ T

x (yS) = (xy)S
(Sx) y = S (xy) for all x, y

∈ T.

If in particular T = KL then we say the subring S is a normal subring of KL.

T

HEOREM

2.3.13: Let C(L) be the center of the loop L and K any field, KL the loop

ring of the loop L over K. KL has a normal subring.

Proof: Obvious form the fact if C(L) is the center of L then KC(L) is a loop ring
contained in KL which is a normal subring of KL.

T

HEOREM

2.3.14: Let L be a loop. S a normal subloop of L. K any field the loop ring

KL has KS to be a normal subring.

Proof: Follows from the fact KS

⊂ KL is such that KS is a normal subring, as S is a

normal subloop of L. Hence the claim.

Now we proceed on to define Smarandache normal subring in a non-associative ring.

D

EFINITION

2.3.8: Let R be a NA-ring. Suppose A be a S-subring of R. We say an

element x

∈ R to be Smarandache normal element (S-normal element) if xA = Ax. If

every x

∈ R is such that xA = Ax we call R a Smarandache normal ring (S-normal

ring) relative to A. Now it is pertinent to mention here that Smarandache normality
is a relative concept, for relative to one S-subring A the ring R may be Smarandache
normal and relative to some other S-subring B of R, the ring R may fail to be a
Smarandache normal.

T

HEOREM

2.3.15: Let R be a S-ring which is normal. If R has S-subring then R is S-

normal. In fact R is S-normal with respect to every S-subring A of R.

Proof: Follows by the very definition.

T

HEOREM

2.3.16: Let R be a NA-ring. If R is a S-normal relative to a S-subring A of

R then R need not in general be a normal ring.

Proof: The reader is advised to obtain one such example.

Thus we see S-normality of R in general need not imply normality of R or R is to be
S-normal relative to every S-subring of R.

EFINITION

2.3.9: Let R be a ring. if for every S-subring A of R is Smarandache

normal relative to A then we call the ring R to be Smarandache strongly normal ring
(S-strongly normal ring).

T

HEOREM

2.3.17: If R is a S-strongly normal ring then R is a S-normal ring and if R

is a S-normal ring then R need not in general be a S-strongly normal ring.

Proof: This is left as an exercise for the reader to prove.

We now proceed on to define the concept of strictly right loop ring and then we will
generalize it to the notion of Smarandache strictly right loop ring.

D

EFINITION

2.3.10: Let R be a non-associative ring. R is said to be a strictly right

ring if the set of all right ideals of R is ordered by inclusion. If all ideals are included
by inclusion we call them as strictly ideal rings.

D

EFINITION

2.3.11: Let R be a non-associative ring. If the set of all S-right ideals of

R are ordered by inclusion then we call R a Smarandache strongly right ideal ring
(S-strongly right ideal ring). In particular if all S-ideals of R are ordered by
inclusion then we call the ring R to be a Smarandache strongly ideal ring (S-strongly
ideal ring).

We do not have any form of relations existing between them but we have several
problems in this direction suggested in the last chapter.

T

HEOREM

2.3.18: Let R be a S-strictly right ideal ring then R is a S-ring.

Proof: Obvious by the very definition of these concepts.

Example 2.3.3: Let Z

= {0, 1} be the prime field of characteristic and let L be a

loop given by the following table:

. e 1 2 3 4 5
e e 1 2 3 4 5
1 1 e 3 5 2 4
2 2 5 e 4 1 3
3 3 4 1 e 5 2
4 4 3 5 2 e 1
5 5 2 4 1 3 e

L be the loop ring of the loop L over the ring Z

. It is left for the reader to find all

the S-right ideals of Z

L, right ideals of Z

L and ideals and S-ideals of Z

The meagerly used concept is the notion of semi-idempotents introduced by [30].

D

EFINITION

[30]: Let R be a ring. An element

α ≠ 0 in R is said to be a semi-

idempotent if and only if

α is not in the two sided ideal of R generated by α

–

α, that

α ∉ R (α

–

α) R or R (α

–

α) R = R, where by R (α

–

α) R we mean the two

sided ideal of R generated by

–

α. So R (α

–

α) R just denotes the ideal

generated by

–

α. We shall denote the set of all semi-idempotents of a ring R by SI

(R).

Example 2.3.4: Let K be any field of characteristic 0. L be a loop given by the table.

. 1 a b c d e
1 1 a b c d e
a a 1 d b e c
b b d 1 e c a
c c b e 1 a d
d d e c a 1 b
e e c a d b 1

The loop ring KL has

and

as some of its nontrivial idempotents in it.

T

HEOREM

2.3.19: Let KL be the loop ring of any loop L over the field K. If

α (≠ 0) is

a semi-idempotent but not a unit in KL, then

α – 1 is not a unit of KL.

Proof: Suppose

α – 1 be unit in KL. Then there is an element β of KL such that (α –

β = 1. Thus we would have α = (α

–

α) β ∈ (α

–

α) KL, but (α

–

α) KL ≠ KL

because

α is a semi-idempotent, hence α(α – 1) = α

–

α is not a unit hence the

result.

Now we define Smarandache semi-idempotents.

D

EFINITION

2.3.12: Let R be a NA-ring. An element

α ∈ R which is a semi-

idempotent is called a Smarandache semi-idempotent (S-semi-idempotent) if in R if
the S-ideal generated by

–

α does not contain α. We denote the collection of S-

semi-idempotents in R by SSI (R).

T

HEOREM

2.3.20: Let R be a NA ring if

α is a S-semi-idempotent then α is a semi-

idempotent of R.

Proof: Direct by the very definition of S-semi-idempotents.

The proof of the following theorem is left as an exercise for the reader to prove.

T

HEOREM

2.3.21: Let R be a ring. If 0

≠ α ∈ R is a semi-idempotent then α in

general need not be a S-semi-idempotent.

Clearly if R is not a S-ring then it is impossible for the element 0

≠ α ∈ R which is a

semi-idempotent; to be a S-semi-idempotent. Now we proceed on to define
Smarandache quasi-regular elements in a loop ring.

EFINITION

2.3.13: Let R be a non-associative ring. We say an element x

∈ R is

called a Smarandache right quasi regular (S-right quasi regular) if there exists y and
z in R such that

x o y = x + y – xy = 0

and

x o z = x + z – xz = 0

but

y o z = y + z – yz

≠ 0

and

z o y = z + y – zy

≠ 0.

Similarly we define Smarandache left quasi regular elements (S-left quasi regular
elements). x will be called Smarandache quasi regular (S-quasi regular) if it is
simultaneously S-left quasi regular and S-right quasi regular.

T

HEOREM

2.3.22: Let R be a NA-ring, if x

∈ R is S-right (left) quasi regular then x is

right (left) quasi regular.

Proof: Follows from the very definitions of these concepts.

T

HEOREM

2.3.23: Let R be a non-associative ring. If x

∈ R is right quasi regular

then x need not in general be S-right quasi regular.

Proof: Let Z

= {0, 1} prime field of characteristic two, L loop given by the

following table:

. e a b c d

e e a b c d
a a e c d b

b b d a e c

c c b d a e

d d c e b a

L be the loop ring. We have x = e + a + b + c

∈ Z

L is right quasi regular for x o (a

+ b) = 0 for x o (a + b) = e + a + b + c + a + b – a – e – d – b – b – c – a – d = 0.

It is easily verified that x is not S-right quasi regular.

D

EFINITION

2.3.14: Let R be a NA-ring. We say a S (left) right ideal is Smarandache

right (left) quasi regular (S- right (left) quasi regular) if each of its elements are S-
right (left) quasi regular. The Smarandache Jacobson radical (S- Jacobson radical)
[SJ(R)] of a ring R is defined as follows: SJ(R) = {a

∈ R / aR is a S-quasi regular

ideal of R}.

P

ROBLEMS

Find the units in Z

(2) which are SNA-units.

Find all quasi regular elements in Z

(2) which are S-quasi regular.

Find all S-units in Z

(8).

Find those idempotents in Z

(6) which are S-idempotents.

Find SJ (Z

(3)).

Find all S-quasi regular elements of Z

(3).

7. Can

(8) have S-right quasi regular elements which are not S-left quasi

regular elements?

8. Find all quasi regular elements of Z

(3) which are not S-quasi regular

elements.

Find all S-semi-idempotents of Z

(8).

10. Determine SSI (ZL

(3)) and SI (ZL

(3)). Do these sets form any proper

algebraic structures?

11.

Is the loop ring Z

(2) a strictly right ring?

12.

Find all the S-ideals of Z

(3).

2.4 Smarandache substructures in loop rings

In this section we study the substructure properties like Marot loop rings, mod p-
envelope of a loop ring, orthogonal ideals, strongly commutative loop ring, inner
commutative loop ring, pseudo commutative loop rings and the lattice of
substructures of loop rings like ideals, subrings, and their Smarandache analogue.
Several interesting results are proved and we see in loop rings several concepts are
not fully analyzed. So these Smarandache analogue is still in a dormant state. Thus in
this section we introduce several new concepts to loop rings.

The concept of Marot loops rings was introduced and studied in [63]. Here we just
recall the concepts and define Smarandache Marot loop rings.

D

EFINITION

[63]: A commutative loop ring RL with identity is a Marot loop ring if

each regular ideal of RL is generated by a regular element of RL, where by a regular
element of RL we mean a non-zero divisor of RL.

T

HEOREM

[63]: Let F be a field. L be the commutative ordered loop without

elements of finite order. Then the loop ring FL is a Marot loop ring.

Proof: Since F is a field and L is a loop which is ordered and has no elements of
finite order we see the loop ring FL has no zero divisors; hence every element is
regular so all ideals in FL will be generated only regular elements. Thus FL is a
Marot ring.

In view of this result we have the following theorem that is left as an exercise for the
reader to prove.

T

HEOREM

[63]: Let F be a field. FL is a Marot loop ring without divisors of zero if

and only if L is a commutative ordered loop without elements of finite order.

T

HEOREM

[63]: If R is a Marot ring and L is a commutative loop; then the loop ring

RL is a Marot ring.

Proof: RL is a commutative loop ring as both R and L are commutative. To prove
RL is a Marot ring we need to show every regular ideal generated by regular
elements of RL. To prove this we shall prove.

a. If a regular ideal I is generated by a regular element of I then I has no proper

divisors of zero.

b. If a regular ideal I is generated by a divisor of zero than I cannot contain

regular elements.

Proof of (a): Let I be generated by a regular element

α. Let β ≠ 0 in I be such that βγ

= 0 (

γ ≠ 0). Now

(

)

∑

βγ













αδ

whether associative or

not we see

α.Σδ

γ = 0, a contradiction to our assumption as α is a regular element of

I. Hence I cannot contain divisors of zero.

Proof of (b): Let I be generated by a divisor of zero say

β, (β ≠ 0, γ ≠ 0) with βγ = 0.

Let

α ∈ I be a regular element of I.

)

(

)

(

)

(

γβ

βδ

αγ

βδ

That is

αγ = 0 a contradiction, so α ∉ I. Hence the theorem.

EFINITION

2.4.1: Let R be a NA ring. We say R is a Smarandache weak Marot ring

(S-weak Marot ring) if R has a commutative S-subring A such that all S-ideals
associated with A are regular. We say R is a Smarandache Marot ring (S-marot
ring) if for every S-subring A of R is commutative; we have all S-ideals associated
with every S-subring A are regular.

T

HEOREM

2.4.1: If R is a S-Marot ring then R is a S-weak Marot ring.

Proof: Direct by the very definition of these concepts.

We leave the following theorems as an exercise for the reader to prove.

T

HEOREM

2.4.2: Every S-weak Marot ring in general need not be a S-Marot ring.

HEOREM

2.4.3: Let R be a S-Marot ring then R need not in general be a Marot

ring.

To prove the two theorems the reader is advised to construct examples. It is also to
be noted even if R is not a commutative ring still R can be a S-Marot ring or S-weak
Marot ring.

The concept of mod-p-envelope of G was studied in [38]. The structure of mod-p-
envelope was studied in case of loop in [62]. It is proved in [62] the mod-p-envelope
of a loop was only a groupoid.

D

EFINITION

2.4.2: Let L be a loop and K be any field. We say L

∗ = 1 + U where U =

{

α = Σ α

∈ KL / Σα

= 0} is the mod p envelope of L.

We study L

∗ in our own way analogous to mod p-envelope of a group G.

Example 2.4.1: Let L be a loop given by the table below and K = {0, 1} be a field of
characteristic two. Then L

∗

, the mod p-envelope of L is a groupoid with (1 + a + b +

c + d)

= (1 + b + a + c + d) and the number of elements in L

∗

is equal to 16.

∗ 1 a b c d
1 1 a b c d
a a d c 1 b
b b 1 d a c
c c b 1 d a
d d c a b 1

KL is the loop ring of L over K. U = {0, 1 + a, 1 + b, 1 + c, 1 + d, a + b, a + c, a + d,
b + c, b + d, c + d, 1 + a + b + c, 1 + a + b + d, 1 + a + c + d, 1 + b + c + d, a + b + c
+ d}. Let L

∗

= 1 + U;

α = 1 + a + b + c + d ∈ L∗ with α

α and 1 ∈ L

∗

, thus L

∗

is a

groupoid with 16 elements in it.

T

HEOREM

[72]: Let L be a commutative finite loop of order 2n in which the square

of every element is one and let K = {0, 1} be the prime field of characteristic two.
Then L

∗

is a loop which is commutative such that square of each element is 1 and

order of L

∗

is 2

2n – 1

Proof: Let L = {a

= 1, a

, …, a

/ a

= 1, a

= a

; i, j = 1, 2, …, 2n}. Now U =

{0, 1 + a

, …, a

2n –1

+ a

, …, sum of 2n elements taken at a time}. Now L

∗

= 1 + U,

so |L

∗

| = 2

2n – 1

. Clearly L

∗

is a commutative loop as L is a commutative loop. So is L

∗

using the fact a

= 1 for i = 1, 2, …, 2n and characteristic of K is 2.

HEOREM

2.4.4: Let L be a loop of order 2n + 1 commutative or otherwise; K = {0,

1} be the prime field of characteristic two. L

∗

is a groupoid of order 2

with a

nontrivial idempotent in it.

Proof: The proof is left as an exercise for the reader to prove.

T

HEOREM

2.4.5: Let L be a finite loop with an element x

∈ L such that x

= 1. Let Z

= {0, 1, 2, …, p – 1} be a prime field of characteristic p, p > 2. Then L

∗

is a groupoid

with a nontrivial idempotent in it and the order of L

∗

is p

n-1

where |L| = n.

Proof: Straightforward, hence left for the reader as an exercise.

T

HEOREM

2.4.6: Let L be a loop of order p + 1 and K = Z

= {0, 1, 2, …, p – 1}

prime field of characteristic p. Then L

∗

is a groupoid with a nontrivial idempotent in

it.

Proof: Left as an exercise for the reader to prove.

Now we proceed on to define Smarandache mod p-envelope of a loop.

EFINITION

2.4.3: Let L be a S-loop and K any field, KL the loop ring of K over the

loop L. The Smarandache mod p-envelope of L (S-mod p-envelope of L) is the mod p
envelope of the S-subloop A of L denoted by S(L

∗

). Thus unlike in the case of loops

where the mod p envelope is unique, in case of S-mod p-envelope of a loop we may
have several S-mod p-envelope depending on the number of S-subloops in L. If L has
no S-subloop and even if L is a S-loop still the S-mod p envelope of L will only the
empty set.

Thus it is an interesting research problem to find the number of S-mod p-envelopes
associated with a loop L.

T

HEOREM

2.4.7: Let L be a loop having no S-subloops or L is not a S-loop and K

any field. Then S (L

∗

) =

φ.

Proof: Straightforward from the very definition of S (L

∗

HEOREM

2.4.8: Let L

(m)

∈ L

where n is a prime. K any field, S (L

∗

) =

φ.

Proof: Clearly L

(m)

∈ L

when n is a prime is a S-loop; but L

(m) has no proper S-

subloop, hence S ((L

(m))

∗

φ.

Now we study the loop rings when the loops are unique product or two unique
product loops.

D

EFINITION

2.4.4: A loop L is called a two unique product loop (t.u.p loop), if given

any two non-empty finite subsets A and B of L with |A| + |B| > 2, there exists atleast
two distinct elements x and y of L that have unique representation in the form x = ab,
y = cd with a, c

∈ A and b, d ∈ B. A loop L is called a unique product loop (u.p.

loop) if, when A and B are non-empty finite subsets of L, then there always exists
atleast one x

∈ L which has unique representation in the form x = ab with a ∈ A and

∈ B.

[52] proved in case of groups t.u.p and u.p are equivalent for groups. In case of loops
it an open question.

T

HEOREM

2.4.9: Let R be any ring with identity and L be a t.u.p. power associative

or diassociative loop. Then the following are equivalent.

i. U (KS) = {

Σ α

g / there exists a

∈ R with Σ α

-1

= 1 and

= 0

whenever gh

≠ 1}, where S is the subgroup generated by single element or a

pair of elements.

ii. R has no nonzero nilpotent element.

Proof: To prove (i) implies (ii) follows from the fact that if R has nilpotent elements
say

γ ∈ R then 1 + γg is a unit in RS. For if γ

= 0 then

(1 +

γg) (1 – γg + γ

–

+ …

± γ

n-1

) = 1 –

γg + γ

– …

± γ

n-1

γg –

+ …

± γ

n-1

± γ

= 1 as

= 0 and all other terms cancel out. To show (ii)

implies (i) the result follows from the fact if R has non-zero nilpotents and p, q

∈ RL

where L is a t.u.p loop. If pq = 1, where p =

Σ α

g and q =

Σ β

h then

= 0

when gh

≠ 0 with some factors using t.u.p we get the result. Hence the claim.

EFINITION

2.4.5: Let L be a loop. We say L is a Smarandache unique product loop

(S.u.p loop) if there exists a S-subloop A of L such that A is a u.p loop. We call the
loop L a Smarandache strongly u.p loop (S-strongly u.p.loop) if every S-subloop A of
L is a u.p subloop of L.

D

EFINITION

2.4.6: Let L be a loop; L is said to be Smarandache t.u.p loop (S-t.u.p

loop) if L has a S-subloop A such that A is a t.u.p subloop of L. If every S-subloop of
L is a t.u.p loop then we say L is a Smarandache strong t.u.p (S-strong t.u.p) loop.

T

HEOREM

2.4.10: If L is a S-u.p loop or S.t.u.p loop then L is a S-loop.

Proof: Obvious by the very definitions.

T

HEOREM

2.4.11: Let L be a S-loop which is a u.p loop or a t.u.p loop If L has no S-

subloops then L is not a S.u.p loop or a S.t.u.p loop.

Proof: It is left as an exercise for the reader to prove.

T

HEOREM

2.4.12: Let L

(m)

∈ L

, if p is a prime; no L

(m) is a S.u.p loop or a S.t.u.p

loop.

Proof: Follows from the fact if p is a prime then L

(m) has no S-subloop. So L

(m)

is not a S.t.u.p loop or a S.u.p loop.

A loop L is Hamiltonian if every subloop is normal. In view of this we define
Smarandache Hamiltonian loop as follows:

D

EFINITION

2.4.7: Let L be a loop if every S-subloop of L is normal then we say L is

a Smarandache Hamiltonian loop (S-Hamiltonian loop).

T

HEOREM

2.4.13: Let L be a finite diassociative Hamiltonian loop which is the

direct product of A

× T × H where A is an abelian group with elements of finite

order, T an abelian group of exponent 2 and H satisfies (x, y, z) = 1 and K any field.
The loop algebra KL is a direct sum of irreducible KA-modules, KT-modules and
KH-modules.

Proof: Since L is a finite diassociative Hamiltonian loop. KA, KT and KH are group
algebras of finite groups A, T and H respectively. By Maschke’s theorem KA is a
direct sum of irreducible KA-modules, KT is the direct sum of irreducible KT-
modules and KH is the direct sum of irreducible KH modules. Hence KL is a direct
sum of irreducible KA, KT and KH-modules. Hence the theorem.

D

EFINITION

2.4.8: Let RL be the loop ring of the loop L over the ring R. Two non-

zero ideals J and I of RL are orthogonal if I.J = {i.j = 0 for all i

∈ I and j ∈ J}. An

ideal I of RL is self orthogonal if I.I = (0).

Example 2.4.2: Let Z

= {0, 1} be the prime field of characteristic two and L be the

loop given by the following table:

∗  1 a b c  d e f  g
1 1 a b c  d e f  g
a a 1 c e  g b d f
b b g 1 d  f  a c e
c c f a 1  e g b d
d d e g b  1 f  a c
e e d f a  c 1 g b
f  f  c e g

b d 1 a

g g b d f a c e 1

Consider the loop ring Z

L. Take two ideals given by I = {0, 1 + a + b + c + d + e +

f + g} and J = {0,

α = Σ m

/ m

∈ L} such that the sum of the coefficients of m

the sum

α is zero which are clearly nontrivial and are such that I.J = {0}. Thus I and

J are orthogonal ideals of Z

L. It is easily verified that I.I = {0}.

Example 2.4.3: Let Z

= {0, 1} be the prime field of characteristic two and L be the

loop given by the following table:

. 1 a b c d
1 1 a b c d
a a 1 c d b
b b d a 1 c
c c b d a 1
d d c 1 b a

L be the loop ring of the loop L over Z

Consider the ideals

I = {0, 1 + a, b + c, b + d, c + d, a + b + c + 1, 1 + a + b + d, 1 + a + d + c} and
J = {0, 1 + b, a + d, c + d, a + c, 1 + a + b + d, 1 + b + c + d, 1 + b + c + d, 1 + b + a
+ c}, K = {ideal generated by I and J}. It is easily verified that I.J

≠ {0}, J.K ≠ {0}

and I.K

≠ {0}.

HEOREM

2.4.14: Let Z

= {0, 1} be the prime field of characteristic two and L be a

finite loop of even order. Then the loop ring Z

L contains non-zero orthogonal

ideals.

Proof: Since |L| = 2n, we have I = {0, 1 + m

+ … + m

} and J = {

α = Σ α

Σ α

= 0 and m

∈ L} clearly I and J are orthogonal ideals.

HEOREM

2.4.15: Let L be a loop of odd order and Z

= {0, 1} be the prime field of

characteristic two, then the loop ring Z

L need not in general contain orthogonal

ideals.

Proof: Left for the reader to prove.

T

HEOREM

2.4.16: Let L be a loop of order n with p/n and Z

be the odd prime field

of characteristic p. Then the loop ring Z

L has nontrivial orthogonal ideals.

Proof: Let L = {1, m

, …, m

} and Z

= {0, 1, 2, …, p – 1}, p > 2 and p/n. Z

L the

loop ring of the loop L over Z

. Consider the non-zero ideals given by I = {0,

γ (1 +

+ … + m

) /

γ = 1, 2, …, p – 1} and J = {α = Σα

Σ α

= 0}; clearly I.J = {0}.

HEOREM

2.4.17: Let L = {1, m

, …, m

} and Z

= {0, 1, …, p – 1}. Z

L the loop ring

of the loop L over Z

. Clearly I = {0,

γ (1 + m

+ … + m

) /

γ = 1, 2, …, p – 1} is a

non-zero ideal of Z

L such that I

= {0}.

The study of orthogonal ideals in loop rings in absent. So we have discussed to a
possible extent the concept of orthogonal ideals. This notion will find its place in
algebraic coding theory. Now we proceed on to define the concept of Smarandache
orthogonal ideals.

D

EFINITION

2.4.9: Let KL be a loop ring of a loop L over the ring K. Let I and J be

two ideals of KL. I and J are said to be Smarandache orthogonal ideals (S-
orthogonal ideals) if the following conditions are satisfied.

i In I.J = {i, j / i

∈ I and j ∈ J} we have atleast a pair of elements i ∈ I and j ∈

J such that i

≠ 0 and j ≠ 0 with i.j = 0.

ii Every one of the pairs in I.J which are such that i.j = 0 are S-zero divisors.

HEOREM

2.4.18: Let KL be the loop ring of the loop L over the ring K. If every pair

of ideals are orthogonal; then also KL need not have Smarandache orthogonal
ideals.

Proof: To prove this it is enough if the reader can show a pair (x, y)

∈ I × J. I and J

ideals of KL with x

∈ I \ {0} and y ∈ J \ {0} with x.y = 0 and x.y is not a S-zero

divisor.

T

HEOREM

2.4.19: Let KL be the loop ring of the loop L over the field K. If KL has

orthogonal ideals and no S-divisors of zero then KL has no S-orthogonal ideals.

Proof: Left for the reader to prove.

Now we proceed on to define a new concept called right commutatively and its
Smarandache analogue in loop rings.

D

EFINITION

2.4.10: Let RL be the loop ring of a loop L over the ring R with unity. If

for every triple

γ, β, α ∈ RL we have (αβ) γ or α(βγ) = α (γβ) or (αγ) β, then we call

the loop ring RL to be a strongly right commutative, i.e., for every triple

α, β, γ ∈ RL

we must have

α (βγ) = α (γβ) or (αγ) β.

b. (

αβ) γ = (αγ) β or α (γβ).

then we say RL is strongly right commutative. The term ‘or’ in general is not used in
the mutually exclusive sense.

T

HEOREM

2.4.20: Every strongly right commutative loop ring RL is commutative.

Proof: Obvious from the fact for every triple

α, β, γ ∈ RL we have αβ = βα, αγ = γα

and

βγ = γβ where one of α or β or γ is assumed to be the unit of the loop ring RL.

Hence every pair is commutative.

T

HEOREM

2.4.21: Let RL be a strongly right commutative loop ring of the loop L

over a ring R. Then the loop L is commutative.

Proof: Obvious from the fact RL is commutative so L

⊂ RL must also be

commutative.

D

EFINITION

2.4.11: On similar lines one can define strongly left commutative loop

rings as (

αβ) γ = (βα) γ or β (αγ) for every α, β, γ ∈ L.

EFINITION

2.4.12: Let RL be the loop ring of the loop L over the ring R. If for every

pair of elements

α, β ∈ R there exists an element γ ∈ R \ {0, 1} such that γ (αβ) or

(

γα) = (γβ)α or γ(βα); then the loop ring RL is said to be right commutative.

Remark:

γ(αβ) = γ(βα) or (γβ)α or (γα)β) = γ(βα) or (γβ)α. Similarly one can define

left commutative loop rings.

T

HEOREM

2.4.22: Let R be a commutative ring with unity and L a non-commutative

loop. If the loop ring RL is a right commutative loop ring then RL has nontrivial
divisors of zero.

Proof: Follows from the fact

γ(αβ) = γ(βα) (when γ ≠ 0 and γ ≠ 1), we have γ(αβ –

βα) = 0. Hence the claim.

T

HEOREM

2.4.23: Let RL be the loop ring of the loop L over the ring R. If L is a

strongly right commutative ring then RL is a right commutative ring.

Proof: Straightforward by the very definition.

T

HEOREM

2.4.24: Let RL be a right commutative loop ring having no zero divisors.

Then RL is commutative.

Proof: Follows from the fact

γ(αβ) = γ(βα); γ ≠ 0, or 1 γ(αβ – γα) = 0. Since RL has

no divisors of zero we have

αβ = βα.

HEOREM

2.4.25: Let RL be a right commutative loop ring then RL is a weakly right

commutative ring.

Proof: Straightforward.

D

EFINITION

2.4.13: Let RL be the loop ring of the loop L over R. RL is said to be

strictly right commutative if for every pair of elements

α, β ∈ RL we have a γ ∈ RL \

{0, 1} such that

γ(αβ) = γ(βα) or (γα)β = (γβ)α.

HEOREM

2.4.26: Every strictly right commutative loop ring is right commutative.

Proof: Straightforward by the definition.

D

EFINITION

2.4.14: Let RL be the loop ring of the loop L over R. We say RL is a

Smarandache strongly right commutative (S-strongly right commutative) if for every
S-subring A of RL; for all

α, β, γ ∈ A we must have

α(βγ) = α(γβ) or (αγ)β or
(

αβ)γ = (αγ)β orα(γβ)

then we say RL is Smarandache strongly right commutative loop ring (S-strongly
right commutative loop ring).

The term ‘or’ in general is not used in the mutually exclusive sense. If in a loop ring
RL we have atleast one S-subloop A which is strongly right commutative then we call
the loop ring RL to be a Smarandache-right commutative (S-right commutative).

T

HEOREM

2.4.27: If the loop ring RL is S-strongly right commutative then RL is S-

right commutative.

Proof: Direct from the very definition.

We proceed on to define inner commutative loops and its Smarandache analogue.

D

EFINITION

2.4.15: Let RL be the loop ring of a loop L over R. We say the loop ring

RL is inner commutative if every subring of RL is commutative.

T

HEOREM

2.4.28: Let R be a field or a commutative ring with 1. L

(m)

∈ L

be a

loop where p is an odd prime. The loop ring RL

(m) is inner commutative.

Proof: Follows from the fact L

(m) has only subloops, which are subgroups of order

2. Hence RL

(m) is inner commutative.

HEOREM

2.4.29: Let R be a field and L be an inner commutative loop. Then the

loop ring RL is an inner commutative loop ring.

Proof: Straightforward from the definitions.

T

HEOREM

2.4.30: Let RL be a commutative loop ring then RL is an inner

commutative loop ring.

Proof: Follows from the very definitions.

All loop rings which are inner commutative need not be commutative. In view of this
we propose the following.

T

HEOREM

2.4.31: An inner commutative loop ring RL in general need not be

commutative.

Proof: By an example ZL

(9) be the loop ring of the loop L

(9) over Z. Clearly

(9) is an inner commutative loop ring which is not commutative.

Now we proceed on to define Smarandache-analogues.

D

EFINITION

2.4.16: Let RL be the loop ring of the loop L over R. We say RL is a

Smarandache inner commutative loop ring (S-inner commutative loop ring) if every
S-subloop of RL is commutative.

Example 2.4.4: Let ZL

(5) be the loop ring of the loop L

(5) over Z. Z

(5) is a S-

inner commutative loop ring.

D

EFINITION

2.4.17: Let RL be the loop ring for ab = ba in RL if we have (a

α)b =

αa) (or = b(αa)) for all α∈ RL, then we say the pair a, b is pseudo commutative.

[or a(

αb) = (bα)a or b(αa)] i.e., (aα)b or a(αb) may be taken as the initial point, for

both may or may not be equal.

We call the loop ring to be a pseudo commutative loop if every pair a, b in L is
pseudo commutative.

T

HEOREM

2.4.32: Let RL be a commutative loop ring then RL, is a pseudo

commutative loop ring.

Proof: Obvious by the very definition.

D

EFINITION

2.4.18: Let RL be a loop ring. We say RL is a Smarandache pseudo

commutative loop ring (S-pseudo commutative loop ring) if RL has atleast one S-
subring A which is pseudo commutative.

We say RL is Smarandache strongly pseudo commutative (S-strongly pseudo
commutative) if every S-subring A or RL is pseudo commutative.

T

HEOREM

2.4.33: Let RL be a S-strongly pseudo commutative loop ring then RL is

S-pseudo commutative.

Proof: Straightforward by the very definition.

T

HEOREM

2.4.34: Let RL be a S-strongly pseudo commutative loop ring; then RL in

general need not be a pseudo commutative loop ring.

Proof: Consider the loop ring Z

(5). Clearly Z

(5) is a S-strongly pseudo

commutative loop ring.

But Z

(5) is not a commutative loop ring. Further Z

(5) is not a pseudo

commutative loop ring.

D

EFINITION

2.4.19: Let RL be the loop ring of the loop L over the ring R. The pseudo

commutator of RL denoted by P (RL) =

〈{p ∈ RL / a (αb) = p (bα) a; a, b, α ∈ RL}〉;

where

〈 〉 denotes the subring generated by all p’s.

Now one of the natural question would be if L is a loop having P (L) to be the
pseudo commutator; find the relation between P (RL) and R (P (L)).

D

EFINITION

2.4.20: Let RL be a loop ring. The Smarandache pseudo commutator (S-

pseudo commutator) subring of RL are defined and denoted by S (P (RL)) =

〈{p ∈

RL / a (

αb) = p (bα)a where a, b α ∈ A; A a S-subring of RL}〉. The subring

generated by p. We may have several or one S-pseudo commutator subring of RL
depending on varying S-subrings A of RL. It may happen even for varying S-subrings
A of RL we may have the same S (P (RL)).

The reader is expected to find interesting and innovative results in this direction. The
major feature about substructures of any algebraic structure is that how does the
lattice of the collection of all substructures look like.

For, in case of groups we see the normal subgroups forms a modular lattice the two-
sided ideals of a ring form a modular lattice.

Now we study what is the lattice structure of the following substructures.

a.  subrings of a loop ring.
b.  S-subrings of a loop rings.
c.  right (left) ideals of a loop ring.
d.  ideals of a loop ring.
e.  S-right (S-left) ideals of loop ring.
f.  S-ideals of a loop ring.

In our opinion such study for loop rings has not been carried out. Some times the
collection of the substructures may be a distributive or a modular lattice.

The more important factor is sometimes the substructures may be a supermodular
lattice. (we say a lattice L is super modular if for all x, y, z, a

∈ L. We have (a ∪x) ∩

∪ y) ∩ (a ∪ z) = a ∪ {(x ∩ y) ∩ (a ∪ z)} ∪ {(x ∩ z) ∩ (a ∪ y)} ∪ {(y ∩ z) ∩ (a

∪ x)} this will be called as the supermodular identity for simplicity of notation we
replace ‘

∪’ by ‘+’ and ‘∩’ by ‘.’; thus the identity reads as (a + x) (a + y) (a + z) = a

+ (xy) (a + z) + xz (a + y) + yz (a + x) for all a, x, y, z in L)

Example 2.4.5: Let Z

= {0, 1} be the prime field of characteristic two and L be a

loop given by the following table:

. 1 a b c d

1 1 a b c d

a a 1 c d b

b b d a 1 c

c c b d a 1

d d c 1 b a

The only ideals of Z

L, the loop ring of the loop L over Z

are

L, {0}, I

= {0, (1 + a + b + c + d)}, I

= {

α = Σ α

Σ α

= 0}. The set of ideals

are {Z

L, {0}, I

, I

is a modular lattice. In fact we can see that this set also forms a supermodular lattice.

Now we can characterize when will the ideals of a loop ring does not form a super
modular lattice. We denote by A

∪ B = A + B, mean only the ideal generated by the

set A

∪ B.

HEOREM

2.4.35: Let L be a finite loop and K any finite field, KL be the loop ring. S

denotes the collection of all ideals of KL. S is a super modular lattice if and only if S
does not contain a set of 4 distinct ideals say A, B, C, D such that A

∪ B = A ∪ C =

∪ D > A and A > B ∩ C, A > B ∩ D, A > D ∩ C.

Proof: By the symbol ‘>’ we mean A

∪ B has more elements than A. Similarly A >

∩ C implies A has more elements than B ∩ C i.e., not only B ∩ C, it is properly

contained in A.

Suppose S contains a set of 4 distinct elements A, B, C, D such that A

∪ B = A ∪ C

= A

∪ D > A and A > B ∩ C, A > B ∩ D, A > D ∩ C then in the equation.

∪ B) ∩ (A ∪ C) ∩ (A ∪ D) = A ∪ [(B ∩ C) ∩ (A ∪ D)] ∪ [(D ∩ B) ∩ (A ∪ C)]

∪ [(C ∩ D) ∩ (A ∪ B)] we get right hand side of the equation to be A ∪ B where as
the left hand side is A. Clearly A

∪ B ≠ A as A and B distinct. Hence S is not a super

modular lattice.

Conversely suppose S is not supermodular to prove S contains a set of 4 distinct
ideals A, B, C, D such that A

∪ B = A ∪ C = A ∪ D > A and A > B ∩ C, A > B ∩

D, A > D

∩ C. If S is not supermodular we have (A ∪ B) ∩ (A ∪ C) ∩ (A ∪ D) >

∪ {(B ∩ C) ∩ (A ∪ D)} ∪ {(D ∩ B) ∩ (A ∪ C)} ∪ {(D ∩ C) ∩ (A ∪ B)}.

{0}

Figure 2.4.1

Put a = a

∪ [(B ∩ C) ∩ (A ∪ D)] ∪ [(B ∩ D) ∩ (A ∪ C)] ∪ {(D ∩ C) ∩ (A ∪ B)]

and b = B, c = C, and d = D. a

∪ b = A ∪ B, a ∪ C = A ∪ C, a ∪ d = A ∪ D. So

∪ b) ∩ (a ∪ c) ∩ (a ∪ d) > a. It is easily verified a > b ∩ c, a > c ∩ d and a > b ∩

d. Hence the claim.

Now we proceed on to see what is structure of strongly modular lattice. We at the
outset say in a lattice L ‘+’ denotes ‘

∪’ and ‘.’ denotes ‘∩’.

EFINITION

2.4.21: Let L be a lattice we say an element a

∈ L is a strongly modular

element of L if (a + b) (a + c) (a + d) (a + e) = a + bc (a + d) (a + e) + bd (a + c)
(a + e) + be (a + b) (a + c) + cd ( a + b) (a + e) + ce (a + b) (a + d) + de (a + b) (a
+ c) for all b, c, d, e

∈ L.

This identity is known as the strongly modular identity. If in a lattice L all elements
satisfy the strongly modular identity then we say L is a strongly modular lattice.

Now our interest is to see when do the collection of all ideals of a loop ring form a
strongly modular lattice.

Example 2.4.6: Let Z

= {0, 1} be the prime field of characteristic two and L be a

loop given by the following table:

. 1 a b c d e f

1 1 a b c d e f

a a f c b e 1 d

b b d f 1 a c e

c c e d a f b 1

d d 1 e f c a b

e e c 1 d b f a

f f b a e 1 d c

L be the loop ring of the loop L over Z

. Ideals of Z

L are {0} = I

, I

= {0, 1 + a +

b + c + d + e + f}, I

= {0,

Σ α

= 0} and Z

L. The collection of ideals be

denoted by S = {I

, I

, Z

L}.

The lattice diagram of S is

which is easily verified to be a strongly modular lattice.

= {0}

Figure 2.4.2

HEOREM

2.4.36: Let L be a finite loop and K any field. KL the loop ring of the loop

L over the field K. S denotes the set of all ideals of KL. S is strongly modular if and
only if there does not exists 5 distinct elements A, B, C, D, E in S such that A

∪ B =

∪ C = A ∪ D = A ∪ E > A and A > B ∩ C, A > C ∩ E, A > D ∩ E, A > B ∩ D, A

> B

∩ E and A > C ∩ D.

Proof: The proof follows by using the strongly modular identity and the mentioned
relations in the theorem.

Now in case of S-ideals and S-subrings the results given in those theorems hold
good. But it is pertinent to mention here that we are not able to find other than the
two theorems mentioned in this section about the S-ideals or ideal or subrings or of
S-subrings.

P

ROBLEMS

Is the loop ring Z

(8) a Marot loop ring? Justify your claim.

Find all the regular elements of the loop ring Z

(3).

Give an example of a loop ring which is a weak Marot loop ring.

4. Find

∗

for the loop ring Z

(3).

Does the loop ring Z

(8) have orthogonal ideals?

6. Can

(8) have S-orthogonal ideals?

Give an example of a strongly right commutative loop ring.

8. Is

(5) inner commutative? Is it S-inner commutative?

Give an example of a pseudo commutative loop ring.

10.

Is the loop ring Z

(3) S-pseudo commutative?

11.

Find the lattice of S-subring of Z

(3).

12.

Find all S-ideals of Z

(3). Does it form a modular lattice?

13.

Find all ideals of Z

(5). Do they form a supermodular lattice?

14.

Give an example of a loop whose S-ideal forms a strongly modular lattice.

15.

Give an example of a loop ring where only S-ideals form a modular lattice
but ideals do not form a modular lattice.

2.5 General properties of Smarandache loop rings and loop rings

The determination of group of units in a group ring is an interesting one to several
researchers. E.G. Goodaire [24 to 29] has studied this problem for loop rings in
1986. The isomorphism problem for group rings was proposed in [32]. He asks if
isomorphism of integral group rings ZH

≈ ZG implies that the groups G and H are

isomorphic. The problem has been settled only for special cases and remains open.
The famous semi simplicity problem of finding necessary and sufficient condition
for J(RZ) = 0 which remains still unsettled even over field of characteristic zero this
has been studied by [24 to 29] in case of alternative loop rings. Very recently it has
been proved that every representation of a semi simple analytic Moufang loop is
completely reducible. Here we derive the Smarandache analogue of all these and
suggest some problems.

EFINITION

2.5.1: An alternative ring is one in which the associator (x,y,z) = (xy)z –

x(yz) is a skew symmetric function of its arguments.

It was proved that if in an alternative ring the subring generated by any pair of
elements is associative.

T

HEOREM

[29]: Suppose ZL is an alternative ring which is not associative and G is

any subgroup of index 2 in L. Then

(i) The

map

∗ : G → G defined by







∉

∈

)

(

)

(

is an involution on G which extends to an involution on the group ring
ZG as follows.

∑













∈

(ii)

Every element in ZL can be written in the form x + yu where x , y
∈ ZG. Multiplication in ZL is given by (x + yu) (z + wu) = (xz +
g

∗

y) + (wx + yz

∗

) u where x, y, z, w

∈ Z G and g

is an element

in Z(G).

(iii) The

map

∗ : ZL→ ZL defined by (x + yu)∗ = x

∗

+ eyu is an

involution on ZL extending, the one on ZG given in (i). An element
α ∈ ZL is central if and only if α

∗

α..

(iv)

The centre of the nucleolus of the ring ZL coincides and are equal
to {x+ yu | x, y Z (RG); ey = y} = {x + yu | x

∈ Z (RG) and ey =

y}.

The very natural study in the Smarandache direction for the above theorem is for the
loop ring ZL where ZL in an alternative ring find Smarandache centre and
Smarandache nucleus of the ring ZL.

Do they concide? We see in case of the loops in L

HEOREM

2.5.1: Let L

(m)

∈ L

, The centre Z (L

(m)) and Nucleus N (L

(m)) are

just {e}.

Proof: Left as an exercise for the reader to prove.

The study of the nucleolus of RL

(m) and centre of R L

(m) where R is a ring is an

important study. Do they coincide? or Are they just equal to R {e} = R? and so on.

Find conditions on the units in ZL

(m) so that they are central units in ZL

(m). Can

S-units be central units in ZL

(m)?

If has been proved by [24].

T

HEOREM

2.5.2: Suppose r is a central unit is an alternative loop ring ZL such that

−

r is trivial . Then ZL is a S-loop ring.

Proof: Given r is central, ey = y. since r is trivial either x =

± g + (1 – e) x

, y = (1 –

e) y

for some x

and y

in the group ring [using results from 26] we have ZL is a S-

loop ring.

It is interesting to note the Zassenhaus conjecture for group rings which is still open
in group rings is also open in loop rings all the more in case of Smarandache loop
rings and loop rings which are S-rings.

Zassenhaus conjecture: Let r

∈ ZG be a normalized unit of finite order. Then there

exists an invertible element

α ∈ QG and an element g ∈ G such that α

-1

α = g.

Here Z denotes the ring of rational integers and Q the field of rationals

∈: ZG → Z

the augmentation map; i.e. map given by

∈ (Σα

g) =

Σα

. If we denote by U(ZG)

the group of units of ZG, then the set V (ZG) = {

α ∈ U (ZG) | ∈ (α) = 1} is called

the group of normalized units of ZG.

The elements of the form

±g, g ∈ G are trivial units of ZG. [26] has established the

alternative analogue of the Zassenhaus conjecture as follows;

T

HEOREM

[25]: Let

τ be a normalized torsion unit in the integral alternative loop

ring ZL of a finite loop L. Then there exists units

∈ Q L and m ∈ L such that

)

(

−

= m.

He has proved that this extended conjecture is at least true in case of these alternative
loop rings, which are not associative. Now we propose a analogue of Zassenhaus
conjecture for S-loop rings and SNA-rings. Let KL be a loop ring, which is a SNA-
ring. Then the augmentation map

∈: K L → K where K is the ring of integers given

)

(

∑

∈

If S U (ZG) denotes the set of S-units of ZG then the set S V (ZG) = {

α ∈ S U (ZG)

∈ (α) = 1} is called the group of Smarandache normalized units (S-normalized units)
of ZG. Smarandache Zassenhaus conjecture. Let r

∈ Z L which is Smarandache

normalized unit in the integral loop ring of a finite loop L. Then there exists

∈ QL and m ∈ L such that α

-1

(

-1

)

= m.

It is still an open problem to solve the Smarandache Zassenhous conjecture for

I. Smarandache loop rings.

II. S- loop rings.

ROBLEMS

1. Is

(3)

≅ ZL

(2)?

2. Can

(3) be S-Isomorphic with Z

(3)? Justify your answer

3. Can we say QL

(m) is S-isomorphic with Z

(m)? Justify your

claim.

Find all S-units in Z

(8).

5. Can

(8) have central units?

Verify Zassenhaus conjecture for the loop ring ZL

(3).

7. Will

(3) satisfy S-Zassenhaus conjecture? Justify.

Chapter 3

GROUPOID RINGS AND SMARANDACHE
GROUPOID RINGS

This chapter has 5 sections. In the first section we introduce the concept of groupoids
and Smarandache groupoids. Section two is devoted to the study and introduction of
groupoid rings and Smarandache groupoid rings. In section three we study special
elements in Smarandache groupoid rings and SNA-groupoid rings. Study of
substructure is carried out in section four. In the final section several new properties
like identities satisfied by Smarandache groupoid rings and other special and new
notions about these groupoid rings are introduced and studied, we define level II
Smarandache groupoid rings and study them.

3.1 Groupoids and Smarandache groupoids

In this section we introduce several of the properties of groupoids and Smarandache
groupoids. Clearly as groupoids are the most generalized algebraic structures viz
loops and semigroups; all loops are groupoids and groupoids in general are not
loops. Likewise all semigroups are groupoids but all groupoids in general are not
semigroups. Thus we study these groupoids and Smarandache groupoids with a main
motivation to construct a new class of non-associative rings called groupoid rings.
Groupoid rings are a generalized class of loop rings. Thus this study of these non-
associative rings and mainly their Smarandache NA rings will reveal a lot of
interesting results. The study of Smarandache groupoids started in the year 2002
[70]. Now we introduce and study Smarandache groupoid rings and groupoid rings.

D

EFINITION

3.1.1: A non-empty set of elements G is said to form a groupoid if in G

is defined a binary operation called the product denoted by

∗ such that a ∗ b ∈ G for

all a, b

∈ G. It is important to mention here that the binary operation ∗ defined on

the set G need not be associative i.e. (a

∗ b) ∗ c ≠ a ∗ (b ∗ c) in general for all a, b, c

∈ G.

Throughout this chapter by a groupoid (G,

∗) we assume that G is non-associative

under the operation

∗.

A groupoid G is said to be a commutative groupoid if for every a, b

∈ G we have a ∗

b = b

∗ a. A groupoid G is said to have an identity element e in G if a ∗ e = e ∗ a = a

for all a

∈ G.

EFINITION

3.1.2: Let (G,

∗) be a groupoid a proper subset H ⊂ G is a subgroupoid

if (H,

∗) is itself a groupoid.

Example 3.1.1: Let R be the set of reals (R, –) is a groupoid, where '–' is the usual
subtraction of reals in R.

Example 3.1.2: Let G be a groupoid given by the following table:

∗

This has H = {a

, a

} and K = {a

, a

} to be subgroupoids of the groupoid (G,

∗). A

groupoid G which has only a finite number of elements in them is called a finite
groupoid. A groupoid which has infinite number of elements is called an infinite
groupoid or a groupoid of infinite order.

D

EFINITION

3.1.3: A groupoid G is said to be a Moufang groupoid if it satisfies the

Moufang identity (xy)(zx) = (x(yz))x for all x, y, z in G.

D

EFINITION

3.1.4: A groupoid G is Bol if it satisfies the Bol identity ((xy)z)y =

x((yz)y) for all x, y, z

∈ G. A groupoid is a P-groupoid if (xy)x = x(yx) for all x, y ∈

G. A groupoid G is right alternative if it satisfies the identity (xy)y = x(yy) for all x, y
∈ G and G is left alternative if (xx)y = x(xy) for all x, y in G. A groupoid G is
alternative if it is both right and left alternative. A groupoid is said to be a Jordan
groupoid if it satisfies the Jordan identity x (x

y) = x

(xy) for all x, y

∈ G.

EFINITION

3.1.5: G be a groupoid, P a non-empty proper subset of G. P is said be

a left ideal of G if

(a)

P is a subgroupoid of G.

(b)

For all x

∈ G and a ∈ P we have x a ∈ P. One can similarly define

a right ideal of the groupoid G. P is called an ideal if P is
simultaneously a left and a right ideal of G.

EFINITION

3.1.6: Let G be a groupoid. A subgroupoid V of G is said to be a normal

subgroupoid of G if

a.  aV = Va
b.  (Vx)y = V(xy)
c.  y(xV) = (yx) V

for all x, y, a

∈ G. A groupoid G is said to be simple if it has no nontrivial normal

subgroupoid.

D

EFINITION

3.1.7: A groupoid G is normal if

1.  x G = G x
2.  G(xy) = (Gx)y
3.  y(xG) = (yx)G

for all x, y

∈ G.

EFINITION

3.1.8: Let G be a groupoid H and K be two proper subgroupoids of G

with H

∩ K = φ or {e}. We say H is conjugate with K if there exists x ∈ H such that

H = xK or Kx ‘or’ in the mutually exclusive sense.

Direct products are defined as in case of other algebraic structures.

A groupoid G has a zero divisor if 0

∈ G and for same 0 ≠ a ∈ G, there exists b ≠ 0

∈ G with a ∗ b = 0. An element a ∈ G is an idempotent of G if a ∗ a = a

= a. Let G

be a groupoid. The center of the groupoid C(G) = {x

∈ G / a x = x a for all a ∈ G}.

We say a, b

∈ G is a conjugate pair if a = bx (or xb for some x ∈ G) and b = ay (or

ya for some y

∈ G). An element a in G is said to be right conjugate to b in G if we

can find x, y

∈ G such that a. x = b and b. y = a (x. a = b and y. b = a). Similarly we

define left conjugate elements of G. One of the drawbacks about groupoids was that
we did not have natural examples of groupoids built using Z

. So we recall the

definition of a new class of groupoids using Z

. [70]

EFINITION

3.1.9: Let Z

= {0, 1, 2, …., n-1}; n

≥ 3, n < α. Define a binary

operation

∗ on Z

as follows. For any a, b

∈ Z

define a

∗ b = (at + bu) (mod n)

where (t, u) = 1 where ‘+’ is the usual addition modulo n. {Z

, (t, u),

∗} or Z

(t, u) is

a groupoid. For varying values of t, u

∈ Z

\ {0} we get a class of groupoids for a

fixed n. This class of groupoids is denoted by Z(n) and all groupoids in this class is
of order n. i.e. Z(n) = { Z

, (t, u),

∗ | for integers t, u ∈ Z

\ {0} with (t, u) = 1}.

Several interesting results can be obtained but the reader is requested to refer [W B V
Pad B].

By Z

∗

(n) we denote the class of groupoids of order n built using Z

with the binary

operation ‘.’ such that for a, b

∈ Z

, a.b = at + bu (mod n) where (t, u) need not

always be relatively prime but t

≠ u, t, u ∈ Z

\ {0}. Clearly the class of groupoids

Z(n) is completely contained in Z

∗

(n). Similarly we define Z

∗∗

(n) to be the class of

groupoids built using Z

under the binary operation ‘o’ defined by a o b = ta + ub

(mod n) for all a, b

∈ Z

with t, u

∈ Z

\ {0}. Thus for varying t and u we get a class

of groupoids for any integer n. We denote this class by Z

∗∗

(n). We have Z(n)

⊆ Z

∗

(n)

⊆ Z

∗∗

(n).

Now the most generalized class of groupoids using Z

denoted by Z

∗∗∗

(n) is built

using t, u

∈ Z

with the binary operation ‘

∗’ such that a ∗ b = ta + ub (mod n). Z(n) ⊂

∗

(n)

⊆ Z

∗∗

(n)

⊆ Z

∗∗∗

(n). Thus we get a natural class of groupoids of order n.

Several interesting properties about them can be had from (WBV) P and B).

Now we proceed on to define Smarandache groupoids and study about them.
Smarandache groupoids were introduced only in the 2001.

D

EFINITION

3.1.10: A Smarandache groupoid (S-groupoid) G is a groupoid which

has a proper subset S, S

⊂ G such that S under the operations of G is a semigroup.

Example 3.1.3: The groupoid (G,

∗) given by the following table:

∗ 0 1 2 3 4 5
0 0 3 0 3 0 3
1 1 4 1 4 1 4
2 2 5 2 5 2 5
3 3 0 3 0 3 0
4 4 1 4 1 4 1
5 5 2 5 2 5 2

Clearly S

= {0, 3} and S

= {1, 4} are semigroups of (G,

∗). A S-groupoid (G, ∗) is

said to be a Smarandache commutative groupoid (S-commutative groupoid) if there
is a proper subset which is a commutative semigroup.

Let (G,

∗) be a groupoid. A non-empty subset H of G is said to be a Smarandache

subgroupoid (S-subgroupoid) if H contains a proper subset K

⊂ H such that K is a

semigroup under the operations of

∗.

HEOREM

3.1.1: If G has a S-subgroupoid then G is a itself a S-groupoid.

Proof: Direct by the very definition.

T

HEOREM

3.1.2: Every subgroupoid of a S-groupoid need not in general be a S-

subgroupoid.

Proof: Follows by counter examples.

D

EFINITION

3.1.11: A Smarandache left ideal (S-left ideal) A of a S-groupoid G

satisfies the following conditions.

1. A is a S-subgroupoid of G.
2. x

∈ G and a ∈ A then x a ∈ A.

Similarly, we can define Smarandache right ideal (S-right ideal), if A is both a S-
right ideal and S-left ideal simultaneously then we say A is a Smarandache ideal (S-
ideal).

Let V be a S-subgroupoid of a groupoid G. We say V is a Smarandache seminormal
groupoid (S-seminormal groupoid) if

(i)

aV = X for all a

∈ G

(ii)

Va = Y for all a

∈ G

where either X or Y is a S-subgroupoid of G but both X and Y are subgroupoids of G.

We say V is a Smarandache normal groupoid (S-normal groupoid) if aV = X and Va
= Y for all a

∈ G where both X and Y are S-subgroupoids of G. Let H and P be any

two subgroupoids of a groupoid G we say H and P are Smarandache semi conjugate
subgroupoids (S-semi conjugate subgroupoids) of G, if

1.  H and P are S-subgroupoids of G.
2.  H = xP or Px or
3.  P = x H or Hx for some x

∈ G.

EFINITION

3.1.12: Let H and P be two subgroupoids of G. We say H and P are

Smarandache conjugate subgroupoids (S-conjugate subgroupoids) of G if

1.  H and P are S-subgroupoids of G.
2.  H = xP or Px and
3.  P = xH or Hx.

EFINITION

3.1.13: Let G be a S-subgroupoid. We say G is a Smarandache inner

commutative groupoid (S-inner commutative groupoid) if every semigroup contained
in every S-subgroupoid of G is commutative.

D

EFINITION

3.1.14: Let (G,

∗) be a S-groupoids. If every S-subgroupoid H of G

satisfies the Moufang identity for all x, y, z in H then G is said to be a Smarandache
strong Moufang groupoid (S-strong Moufang groupoid). If there exists at least one
S-subgroupoid H of G which satisfies the Moufang identity then we call G a
Smarandache Moufang groupoid (S-Moufang groupoid).

It is easily verified that all S-strong Moufang groupoids are S-Moufang groupoid,
whereas S-Moufang groupoids are not in general S-strong Moufang groupoids.

On similar lines one can define Smarandache Bol groupoid, Smarandache strong Bol
groupoid, Smarandache alternative (right/left) groupoid, Smarandache strong
alternative (right/left) groupoid and Smarandache strong P-groupoid and
Smarandache P-groupoid. For more about groupoids and S-groupoids please refer [4,
5, 67, 70].

P

ROBLEMS

Find an example of a groupoid of order 7, which is not a S-groupoid.

2. Is

(3,2) a S-groupoid?

Give an example of a S-normal groupoid.

Give an example of a normal groupoid which is not a S-groupoid.

Does there exist a S-strong Bol groupoid of order 10?

Give an example of a Bol groupoid, which is not a S-Bol groupoid.

Will a S-Moufang groupoid be a Moufang groupoid?

Show by an example that a Bol groupoid need not be a S-Bol groupoid.

Find a S-groupoid which is simple and of order 8.

10.

Show by example all simple groupoids need not be S-simple groupoids.

3.2 Groupoid rings and Smarandache groupoid rings

In this section we introduce the concept of groupoid rings and Smarandache
groupoid rings. The study of groupoid rings is very recent. Groupoid rings are
constructed analogous to group rings where groups are replaced by groupoids. Rings

are always assumed to be commutative with unit. Further groupoid rings leads to a
collection of non-associative rings.

These non-associative rings may or may not contain identity. Using different types of
groupoids we can have non-associative rings which are commutative or non-
commutative ring of infinite or finite order. Thus using groupoid rings we get a new
class of non-associative rings which are a generalized class of loop rings for every
loop ring is a groupoid ring. Study of loop rings are carried out in chapter 2.
Throughout this section, G will denote a groupoid may or may not contain 1. R is a
commutative ring with 1 or a field.

D

EFINITION

3.2.1: Let G be a groupoid and R a ring or a field. The groupoid ring

RG consists of all finite formal sums of the form

Σ r

(i – running over a finite

number) where r

∈ R and g

∈ G satisfying the following conditions.

∑

⇔

for i = 1, 2, 3,…., n.

























∑

)

(

∑

where

)

(

)

(

























∑

where

Σ α

4. r

= m

for all r

∈ R and m

∈ G.

∑

)

(

for r

, r

∈ R. Since I ∈ R and m

∈ G, we have G =

1. G

⊆ RG and R ⊆ RG if and only if G has identity otherwise R ⊄ RG.

Clearly RG is a non-associative ring with 0

∈ R as its additive identity. The

groupoid ring RG is an alternative ring if (xx) y = x(xy) and x(yy) = (xy)y for all
x, y

∈ RG.

Let RG be a ring and let x, y

∈ RG. A binary operation known as the circle operation

denoted by ‘o’ is defined by x o y = x + y – xy. The concepts of right (left) quasi
regular elements, the Jacobson radical J(RG) of RG are defined as in the case of any
non-associative ring.

Example 3.2.1: Let Z

= {0,1} be the prime field of characteristic two and G be a

groupoid given by the following table:

. a

G is the groupoid ring given by Z

G = {0, a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

}. |Z

G| = 8 and Z

G is a non-commutative and non-associative ring. We have

α = a

+ a

to be an idempotent for

= (a

+ a

)

= a

+ a

. Clearly

G has no unit element. This is the smallest non-associative ring known to us

without having unit element. We define the concepts of subring and ideals.

D

EFINITION

3.2.2: Let RG be a groupoid ring. We call a subring V of RG to be a

normal subring if

(1) aV = Va
(2) (Vx)y = V(xy)
(3) y(xV) = (yx)V

for all x, y, a

∈ G.

The groupoid ring is simple if it has no normal subrings. We call RG a normal
groupoid ring if

(a) x(RG) = (RG)x
(b) RG(xy) = ((RG)x)y
(c) y(x(RG)) = (yx) (RG)

for all x, y

∈ RG.

EFINITION

3.2.3: Let RG be a groupoid ring of a groupoid G over the ring R. The

centre of RG = C(RG) = {x

∈ RG/ xa = ax for all a ∈ RG}.

EFINITION

3.2.4: Let RG be a groupoid ring. We say a, b

∈ RG is a conjugate pair

if a = bx(or xb) for some x

∈ RG and b = ay (or ya for some y ∈ RG).

We define an element a

∈ RG to be right conjugate in RG if we can find x, y ∈ RG

such that ax = b and by = a (xa = b and yb = a). Similarly we define left conjugate.

We define a groupoid ring RG to be a Moufang ring if the Moufang identity (

αβ)

(

γα) = (α(βγ))α is satisfied for all α, β, γ ∈ RG.

Similarly, we define P-ring, Bol ring and right (left) alternative ring in case of
groupoid rings RG.

The study of these rings mainly depends on the structure of the groupoid G but not
totally on the groupoid G. We can only prove if RG is a Moufang (Bol, P or
alternative) groupoid ring then G is a Moufang groupoid (Bol groupoid, P-groupoid
or alternative groupoid). Now we proceed on to define Smarandache groupoid rings
and Smarandache non-associative rings.

D

EFINITION

3.2.5: Let RG be a groupoid ring; RG is a Smarandache groupoid ring

(S-groupoid ring) if and only if G is a S-groupoid.

EFINITION

3.2.6: Let RG be a groupoid ring; RG is a SNA-ring if and only if RG

has a proper subset which is an associative ring under the operations of RG.

Remark: If RG is a groupoid ring and G is a groupoid with unit then this groupoid
ring is always a SNA-ring. It is inevitable to note that it is not very easy to construct
groupoids with identity. For when we define groupoids with identity we do so by
only adjuring an element with G, but this adjoined element usually is not well
adjusted with G so we in all cases do not get all groupoid rings to be SNA-rings as
most of the groupoids in the new classes of groupoids do not have identity 1 in them.
Thus in all cases we have groupoid rings, which are neither SNA-rings nor S-
groupoid rings. All S-groupoid rings are SNA-rings, we prove this in the following
theorem.

T

HEOREM

3.2.1: Let RG be a groupoid ring which is a S-groupoid ring then RG is a

SNA-ring.

Proof: By the very definition of the S-groupoid ring we see G is a S-groupoid so G
has a proper subset H which is a semigroup. Now consider the semigroup ring RH.
Clearly RH

⊂ RG and RH is an associative ring so RG is a SNA-ring.

HEOREM

3.2.2: A groupoid ring RG which is a SNA-ring need not in general be a

S-groupoid ring.

Proof: Suppose G is a groupoid which is not a S-groupoid with unit 1. Then in the
groupoid ring R

⊂ RG is an associative subring but as G is not a S-groupoid, RG is

not a S-groupoid ring.

Thus we see the class of S-groupoid rings contains completely the class of groupoid
rings which are SNA-rings.

T

HEOREM

3.2.3: Let RG be the groupoid ring. If g

∈ G is an idempotent of G and G

has unit, then the groupoid ring has idempotents.

Proof: Obvious from the fact g

= g for g

∈ G and 1 ∈ RG as 1 ∈ G so g ∈ RG

which is an idempotent

Remark: If g

= g in RG then we have g (g – 1) = 0 is a divisor of zero. But the more

interesting part of the question is that can every RG have S-idempotents for S-
idempotents are defined in non-associative rings only as in the case of associative
rings. For the existence of S-idempotents does not demand the groupoid ring to a S-
groupoid ring or a SNA-ring, what we need is the existence of idempotents in the
groupoid ring.

T

HEOREM

3.2.4: Let RG be a groupoid ring. If the groupoid G has S-subgroupoids

then RG has S-subrings.

Proof: Follows from the very definition of S-subgroupoids and S-subrings. In fact
we can say RG has S-groupoid subrings.

Example 3.2.2: Let G be a groupoid with identity given by the following table:

. 1 a

1 1 a

1 a

= {0,1} the prime field of characteristics two. Z

G be the groupoid ring.

G = {0, 1, a

, a

, 1+ a

, 1 + a

, a

+ a

, a

+ a

, a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, a

+ a

, 1 + a

+ a

G| = 16. We have zero divisors in Z

G, for (1 + a

)

= 0 = (1 + a

)

= (1 + a

)

= (1

+ a

)

= 0. (a

+ a

)

= 0, (a

+ a

)

= 0 and (a

+ a

)

= 0 and (1 + a

+ a

)

1 = (1 + a

+ a

)

= (1 + a

+ a

)

= (a

+ a

)

In view of this we have the following theorem.

T

HEOREM

3.2.5: Let Z

= {0, 1} be the prime field of characteristic two and G be a

commutative groupoid with unit of even order in which every g

∈ G is such that g

1. Then Z

G = K

∪ H where K consist of nilpotent element of order two and H units

of a special form x

= 1 for all x

∈ H.

Proof: Z

H = K

∪ H where K = {Σ α

∈ Z

and g

∈ G with Σα

= 0} and H =

{

Σ α

∈ Z

and g

∈ G with Σα

= 1}. It is given for all g

∈ G, g

= 1 and G is

commutative and order of G is even. Hence for any h

∈ H we have

∑

where n is odd i.e. h = g

+ … + g

; h

= 1 using g

= 1 and g

= g

. Similarly if

∈ K then we have k

= 0. Hence the claim. If G is of odd order we have such a

division only when G is a commutative groupoid with 1 and for every g

∈ G we have

= 1. In Z

G the groupoid ring we have Z

G = H

∪ K as mentioned in the theorem

and H

∩ K = φ.

It is nice to note that H is not even closed under addition where as K is a subring
under the operations of Z

EFINITION

3.2.7: We say the groupoid ring RG is a Smarandache commutative (S-

commutative) if RG has a proper Smarandache subring (S-subring) which is
commutative.

We say RG is Smarandache strongly commutative (S-strongly commutative) if every
proper S-subring of RG is commutative.

Remark: We see even if RG is S-strongly commutative ring still RG need not be
commutative. Further if RG is S-strongly commutative then it is S-commutative.

Suppose RG is a commutative groupoid ring which has no S-subrings then RG is not
S-strongly commutative or S-commutative.

So for S-commutative to exist RG should posses S-subrings which are commutative
even if RG is a commutative groupoid ring still RG need not be S-commutative. We
as in case of S-seminormal groupoid define S-seminormal groupoid ring and S-
normal groupoid rings.

D

EFINITION

3.2.8: Let RG be a groupoid ring. We call a subset W of RG to be a

Smarandache seminormal subring (S-seminormal subring). If

1. W is a S-subring of RG.
2. aW = X for all a

∈ RG.

3. Va = Y for all a

∈ RG

where either X or Y is a S-subring of RG and X and Y are both subrings. We say W is
Smarandache normal subring (S-normal subring) if aV = X and Va = Y for all a

∈

RG where both X and Y are S-subrings of RG.

Result: Every S-normal subring is a S-seminormal subring of RG. Here we define
the concept of n-ideal groupoid rings.

D

EFINITION

3.2.9: Let RG be a groupoid ring we say RG is a n-ideal groupoid ring

if for n-ideals S

, S

, …, S

of RG there exists exactly n-distinct elements x

, x

, …, x

∈ RG \ {S

∪ S

∪ … ∪ S

} such that

〈x

∪ S

∪ …∪ S

〉 = 〈x

∪ S

∪ ….

∪ S

〉 = … = 〈x

∪ S

∪ … ∪ S

〉 where 〈 〉 denotes the ideal generated by x

∪

∪ S

∪ … ∪ S

for 1

≤ i ≤ n.

EFINITION

3.2.10: Let RG be a groupoid ring. S

, S

, …, S

be any n S-ideals of

RG. We say RG is a Smarandache n-ideal ring (S-n-ideal ring) if there exists n
distinct elements x

, … x

in RG \ {S

∪ S

∪ … ∪ S

} such that

〈 x

∪ S

∪…∪

〉 = 〈 x

∪ S

…

∪ S

〉 = … = 〈 x

∪ S

∪ … ∪ S

〉 where 〈 〉 denotes the

S-ideal generated by x

∪ S

∪ …∪ S

for 1

≤ I ≤ n.

It is pertinent to note that n can be any integer value 2 or 3 or 4 or … Depending on
the groupoid G and on the ring R we find whether groupoid ring RG is a S-n-ideal
ring or n-ideal ring for some suitable integer n.

Example 3.2.3: Let Z

= {0, 1} be the prime field of characteristic two and G be a

groupoid given by the following table:

∗

e a

e e a

e a

G be the groupoid ring, K

= {0, 1 + a

, 1 + a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, 1 + a

+ a

, 1 +

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, a

+ a

, 1 + a

+ a

, 1 + a

+ a

, 1+ a

+ a

}, is an ideal of Z

= {0, 1 + a

+ a

} is also an ideal of Z

G. Though there are 2 ideals

in Z

G still it is not an n-ideal ring for any n as K

⊂ K

Using the notion of semi-ideals in groupoid we define Smarandache semi-ideals and
semi-ideals and Smarandache semi-ideals in groupoid rings.

D

EFINITION

3.2.11: Let G be a groupoid. A non-empty subset S of G is said to be a

left semi-ideal of G if

(1) S is closed under the operations of G
(2) x

∈ S for all x ∈ G and s ∈ S.

Similarly we can define right semi-ideals of G. We call S a semi-ideal if it is both
right and left semi-ideal of G.

Example 3.2.4: G be a groupoid given by the following table:

∗ a b c
a b c a
b a b c
c c c c

S = {b, c} is a semi-ideal of G. We define the concept called the strong left (right)
semi-ideal of a groupoid G.

D

EFINITION

3.2.12: Let G be a groupoid. A closed subset S of G is said to be strong

left (right) semi-ideal of G if (xy)s

∈ S (s(xy) ∈ S) for every pair x, y ∈ G and s ∈ S.

HEOREM

3.2.6: If G is a groupoid with S a strong semi-ideal of G, then S is a semi-

ideal of G.

Proof: Straightforward by the definition.

D

EFINITION

3.2.13: Let G be a groupoid. A non-empty subset S of G is said to be a

Smarandache left semi-ideal (S-left semi-ideal) of G if

(1) S is S-subgroupoid of G
(2) x

∈ S for all x ∈ G and s ∈ S

Similarly we define Smarandache right semi-ideal (S-right semi-ideal) of G. If S is
simultaneously both a Smarandache left and right semi-ideal of G then we call S a
Smarandache semi-ideal (S-semi-ideal) of G.

D

EFINITION

3.2.14:

Let G be a groupoid, let A be a non-empty subset of G;

(1) A is a S-subgroupoid of G
(2) For every pair x, y

∈ G and s ∈ G, (xy) s ∈ S.

Then S is said to be a Smarandache strong left semi-ideal (S-strong left semi-ideal)
of G. If s(xy)

∈ S for every s ∈ S and x, y ∈ S we say S is a Smarandache strong right

semi-ideal (S-strong right semi-ideal) of G.

If S is simultaneously both a S-strong right semi-ideal and S-strong left semi-ideal
then we call S a Smarandache strong semi-ideal (S-strong semi-ideal) of G.

Several interesting results about groupoids and its Smarandache properties can be
derived. As the main aim of this book is the study of Smarandache non-associative
rings we do not recall or obtain properties about groupoids or S-groupoids in this
book. Now we proceed on to define Smarandache generalized semi-ideals in
groupoid rings and generalized semi-ideals of groupoid rings.

D

EFINITION

3.2.15: Let G be a groupoid, R any field or a commutative ring with 1.

The groupoid ring RG of the groupoid G over R. Let V be a non-empty subset of RG.
V is said to be a generalized left semi-ideal of RG if V is closed under addition and
x

∈ V for every s ∈ V and x ∈ RG. Similarly we can define generalized right semi-

ideals of RG. If V is simultaneously a generalized right semi-ideal and left semi-ideal
of RG then we call RG a generalized semi-ideal.

D

EFINITION

3.2.16: Let RG be a groupoid ring. We say a non-empty subset V of RG

is a Smarandache generalized left semi-ideal (S-generalized left semi-ideal) of RG if

(1) V is a S-semigroup under ‘+’.
(2) For all s

∈ V and x ∈ RG we have x

∈ RG.

We define similarly Smarandache generalized right semi-ideal (S-generalized right
semi-ideal) of RG. If V is simultaneously both a S-generalized right and left semi-
ideal of RG then we call V a Smarandache generalized ideal (S-generalized ideal) of
RG.

T

HEOREM

3.2.7: Every generalized semi-ideal of a groupoid ring in general need

not be a S-generalized semi-ideal of the groupoid ring of RG.

Proof: We prove this by an example. Consider the groupoid ring Z

G where G is of

finite order say G ={g

, …, g

} such that g

+ …+g

) = g

+ … + g













∑

is a generalized semi-ideal of Z

G, but clearly V is not a S-generalized semi-ideal of

HEOREM

3.2.8: Let Z

G be the groupoid ring where G is a commutative groupoid

with 1 such that g

= 1 for every of g

∈ G. Then Z

G has

(1) S-generalized semi-ideal.
(2) Generalized semi-ideals which is not a S-generalized semi-ideal.

Proof: Consider the groupoid ring Z

G. Let V = {

Σα

= 0}. Clearly it is

verified that V is a S-generalized semi-ideal of Z

G. Now K = {0, 1 + g

+ … + g

}

is a generalized semi-ideal which is not a S-generalized semi-ideal of Z

ROBLEMS

Find normal subrings and S-normal subrings if any in the groupoid ring ZG
where G is the groupoid from Z

∗∗∗

(n).

2. Find S-seminormal subrings and seminormal subrings in the groupoid ring

G where G

∈ Z

∗∗

(n).

Find all units and idempotents in Z

G where G

∈ Z

∗∗

(7).

Find a conjugate pair in Z

G, G

∈ Z

∗

(8).

Give an example of a groupoid ring RG which is S-Moufang.

Find an example of a S-Bol groupoid ring.

Give an example of a S-groupoid ring.

Give an example of a groupoid ring which is not a S-groupoid ring.

Give an example of a groupoid ring which has S-generalized semi-ideals.

10.

Give an example of a groupoid ring, which has no S-generalized semi-ideals.

11.

Give an example of a groupoid ring, which has S-normal subring.

12.

Find for the groupoid ring which has

(1) S-5-ideal ring.
(2) 5-ideal ring.

3.3 Smarandache special elements in Groupoid rings

In this section we specially study the properties of not only elements like units
idempotents etc. but also special propertied rings like E-rings, p-rings, pre J-rings,
zero square rings, quasi commutative rings, pseudo commutative rings etc and
specially their Smarandache analogue. Such study in case of groupoid rings in totally
absent.

The concept of S-units, S-zero divisors and S-idempotents for any non-associative
ring is also the same as that of the associative rings. So we request the reader to refer
[68]. Study of such concepts for loop rings have been carried out we in this section
apart from study the existence of units, S-units, zero divisors, S-zero divisor,
idempotents and S-idempotents in groupoid rings and S-groupoid rings we study and
introduce concepts like normal elements, quasi regular elements, semi idempotents,
pseudo commutative elements and their Smarandache analogue.

Example 3.3.1: Let Z

= {0, 1} be the prime field of characteristic two. G be a

groupoid given by the following table:

. e a

e e a

e a

Take a = (1+a

+ a

) and b = (1+a

). Clearly ab = 0. Choose x, y, x

≠ y such that

x = a

+ a

and y = a

+ a

a . x = 0 and by = 0 but

x . y =

+ a

) (a

+ a

)

+ a

= a

+ a

+ 1

≠ 0.

Thus we see Z

G has S-zero divisors. In view of this example we have the following

theorem.

T

HEOREM

3.3.1: Let Z

= {0, 1} be the field and G be a finite commutative groupoid

with 1 of even order such that g

= 1, for every g

∈ G. The groupoid ring Z

G has S-

zero divisors.

Proof: Let G = {1, g

,… , g

}, where n is odd so |G| is even. Let a = 1+ g

+ … + g

and b = 1 + g

be in Z

G. Clearly ab = 0. Choose x = g

+ g

and y = g

+ g

such that

ax = 0 and by = 0 with xy

≠ 0. Thus the groupoid ring has S-zero divisors.

Example 3.3.2: Let Z

= {0,1} and G be a groupoid given by the following table:

∗

e a

e e a

e a

G is the groupoid ring. (1 + a

+ a

)

= 1. For a

, a

∈ G ( a

≠ a

) and (a

+ a

)

1 for a

, a

∈ G, i ≠ j, j ≠ k and i ≠ k. Similarly (1 + a

+ a

) = 1. a

, a

are distinct elements of G, also (a

+ a

)

= 1 where a

, a

are 5 distinct elements of G. Now,

(

)

∑

Thus we have units. The study of how many of these units happens to be S-units is
an interesting one. Further we see every S-unit is a unit but all units in general are
not S-units.

Now we proceed on to define pseudo commutative elements.

D

EFINITION

3.3.1:

Let RG be non-commutative groupoid ring that is, the groupoid G

is non-commutative. A pair of distinct elements x, y

∈ RG different from the identity

of RG which are such that xy = yx is said to be a pseudo commutative pair of R if
(xa)y or ((ya)x) for all a

∈ RG. If in the groupoid ring RG every commutative pair

happens to be a pseudo commutative pair of RG then we say RG is a pseudo
commutative ring. Clearly if G is a commutative groupoid the groupoid ring RG is a
pseudo commutative ring.

Note: For us to define a pseudo commutative pair in the groupoid ring what we need
is that if xy = yx. We can have at least one of the equalities to be true for all a

∈ RG

is (xa)y = y(ax) or = (ya)x or x (ay) = (ya) x or y (ax).

Now we define Smarandache pseudo commutative groupoid ring.

D

EFINITION

3.3.2:

Let RG be a groupoid ring. Let A be a S-subgroupoid ring of RG.

A pair of elements x,y

∈ A which are such that xy = yx is said to be a Smarandache

pseudo commutative pair (S-pseudo commutative pair) of RG if x(ay) = y(ax) or (ya)
x or (xa) y = y (ya)x) or y (ax) for all a

∈ A. If in the S-subgroupoid ring A every

commutative pair happens to be a S-pseudo commutative pair of A, then A is said to
be a Smarandache pseudo commutative groupoid ring (S-pseudo commutative
groupoid ring).

T

HEOREM

3.3.2: Let RG be a groupoid ring. If RG is a S-pseudo groupoid ring then

RG is a SNA-ring.

Proof: Follows from the fact that RG is a S-pseudo commutative ring then by the
very definition RG is a SNA-ring.

D

EFINITION

3.3.3: The groupoid ring RG is strongly regular if for every x, y

∈ RG

we have (xy)

= xy for n = n(x,y) > 1, then R is called a strongly regular ring. We

say the groupoid ring RG is a Smarandache strongly regular ring (S-strongly
regular ring) if RG contains a SNA-subring A such that for every x, y

∈ A we have

(xy)

= xy for some integer n = n (x, y) > 1.

It is left as an exercise for the reader to prove.

T

HEOREM

3.3.3: If RG is a S-strong regular ring then RG is a SNA-ring.

Such type of study has not been carried out by any researcher in case of any of the
non-associative rings be it loop rings, groupoid rings, Lie rings or Jordan rings. Now
we proceed on to define quasi commutatively in case of groupoid rings RG and their
Smarandache analogue.

D

EFINITION

3.3.4:

Let RG be a groupoid ring we say RG is quasi commutative if

ab= b

a for every pair of elements a, b

∈ R and t > 1.

HEOREM

3.3.4: Let RG be a groupoid ring, if RG is quasi commutative then every

pair of elements in the groupoid G satisfies ab = b

a, t > 1.

Proof: Clearly given in the groupoid ring RG we have ab = b

a, t > 1 for every a, b

∈

RG, further we know I.G

⊆ RG so in G we have ab = b

a for every pair a, b

∈ G; t >

1.

Remark: If G is a groupoid with 1 and if we have RG to be quasi commutative then,
R is quasi commutative. If G has no unit element 1 then we may not be in a position
to say even if RG in quasi commutative so is R.

T

HEOREM

3.3.5: Let RG be a quasi commutative groupoid ring. Then for every pair

of elements a, b

∈ RG there exists s ∈ RG such that a

b = bs

provided (b.b

γ-1

) a =

b(b

γ-1

a).

Proof: RG is a quasi commutative groupoid ring, so ab = b

a for every pair of

elements a, b

∈ RG, γ > 1. Now ab = b

a so a

b = ab

a = (ab) b

γ-1

a = (b

a) (b

γ-1

a) = b

γ-1

a) (b

γ-1

a) = bs

(using b(b

γ-1

a) = (b. b

γ-1

)a where s

= (b

γ-1

EFINITION

3.3.5: Let RG be a groupoid ring. We say RG is a Smarandache quasi-

commutative ring (S-quasi-commutative ring) if for any S-subring A of RG we have
ab = b

a for every a, b

∈ A, γ > 1.

HEOREM

3.3.6: If RG is a S-quasi commutative ring then RG is a SNA-ring.

Proof: Obvious by the very definitions.

The reader is expected to prove the following theorem.

T

HEOREM

3.3.7:

Every SNA-ring in general need not be S-quasi commutative.

EFINITION

3.3.6: An element x of the groupoid ring RG is called semi nilpotent if x

– x is a nilpotent element of RG. If x

– x = 0 we say x is a trivial semi nilpotent.

HEOREM

3.3.8: If x is a nilpotent element of the groupoid ring RG then x is a semi

nilpotent element of RG.

Proof: Straightforward from the fact x

= 0 so x

– x = -x. so (-x)

= 0, hence our

claim.

EFINITION

3.3.7: Let RG be a groupoid ring. An element x

∈ RG is a Smarandache

semi nilpotent (S-semi nilpotent) if x

– x is S-nilpotent. An element 0

≠ x ∈ RG is

said to be Smarandache nilpotent if x

= 0 and there exists a y

∈ R \ {0, x} such that

y = 0 or yx

= 0, r, s > 0 and y

≠ 0 for any integer m > 1.

EFINITION

3.3.8: A groupoid ring RG is reduced if R has no non-zero nilpotent

elements.
We say RG is Smarandache-reduced (S-reduced) if RG has no S-nilpotents.

The following theorem is straightforward and hence left for the reader to prove.

T

HEOREM

3.3.9: Let RG be a reduced groupoid ring then RG is a SNA-reduced ring.

If R is a SNA-reduced ring then RG need not in general be a reduced ring.

D

EFINITION

3.3.9: A groupoid ring RG is a zero square ring if x

= 0 for all x

∈ RG.

EFINITION

3.3.10: A groupoid ring RG is a Smarandache zero square ring (S-zero

square ring) if RG has a S-subring which is a zero square ring.

Example 3.3.3: Let Z

= {0, 1} be the prime field of characteristic two and G be a

commutative groupoid with unit given by the following table:

. 1 a

1 1 a

1 a

Clearly

A =

{

}

with

also

can

∑

∈

∑

is a S-subgroupoid ring which is a S-zero square ring as x

= 0 for all x

∈ A.

Thus we have the following nice theorem.

T

HEOREM

3.3.10: Let Z

= {0,1} be the prime field of characteristic two and G be a

commutative groupoid of even order with 1 such that

g = 1 for every g

∈ G. Then

the groupoid ring Z

G is a S-zero square ring.

Proof: Set U =

{

}

with

∑

∈

∑

; U is such that every element x is of

the form x

= 0. U is a S-subgroupoid ring. Hence Z

G is a S-zero square ring.

HEOREM

3.3.11: Let RG be a groupoid ring such RG is a S-zero square ring then

RG need not be a zero square ring.

Proof: We prove this by the following example. Consider the groupoid ring RG
given in the previous theorem; clearly in RG we have elements x

∈ RG say x = g ∈

RG, g

∈ G such that g

= 1. Hence RG is a zero square ring.

EFINITION

3.3.11: Let RG be a groupoid ring. RG is called an inner zero square

ring if every proper subring of RG is a zero square ring.

We call the groupoid ring RG to be Smarandache inner zero square ring (S-inner
zero square ring) if every proper S-subring of RG is a zero square ring.

Several interesting results can be got from these definitions.

D

EFINITION

3.3.12: We call the groupoid ring RG to be a Smarandache weak inner

zero square ring (S-weak inner zero square ring) if RG has at least a S-subring A

⊆

R G such that a subring B of A is a zero square ring.

D

EFINITION

3.3.13: A groupoid ring RG is said to be p-ring if x

= x and px = 0 for

every x

∈ R. We say the groupoid ring RG is said to be Smarandache p-ring (S-p-

ring) if RG is a S-ring and RG has a subring P such that x

= x and px = 0 for every

∈ P.

HEOREM

3.3.12: Let G be a groupoid with nilpotent elements and R any ring. The

groupoid ring RG is not a p-ring.

Proof: Straightforward from the very definition.

D

EFINITION

3.3.14: Let RG be a groupoid ring. R G is said to be a S-p-ring if RG is

a SNA-ring and has a subring A such that A is a p-ring.

Thus the following theorem is direct hence left for the reader to solve.

T

HEOREM

3.3.13: If RG is a groupoid ring which is a p-ring then RG is a S-p-ring.

HEOREM

3.3.14: Every Smarandache p-ring RG need not in general be a p-ring.

Proof: We prove by an example. Take RG to be a groupoid ring where R is a p-ring
and G is a groupoid with nontrivial nilpotents. Then the groupoid ring RG is a S-p-
ring and not a p-ring. Hence the claim.

Now we proceed on to define a particular case of p-rings in case of associative rings
they are termed as E-ring, So far they have not extended this notion to non-
associative rings. Further this notion cannot be defined on Lie rings which also forms
a class of non-associative rings. We can define the concept of E-rings for loop rings
and groupoid rings that too only when we take the rings R to be of characteristic two
or the prime field of characteristic 2.

D

EFINITION

3.3.15: Let RG be a groupoid ring. We call RG an E-ring if x

= x and

2x = 0 for every x in RG and n a positive integer. The minimal such n is called the
degree of the E-ring.

We have the following nice theorem from the definition which we leave as an
exercise to the reader.

T

HEOREM

3.3.15: Let RG be a E-ring then we have characteristic of R is 2.

EFINITION

3.3.16: Let RG be a groupoid ring. RG is said to be a Smarandache E-

ring (S-E-ring) if RG has a proper subset A which is SNA-subring of RG and A is a
E-ring.

Thus we have the following.

T

HEOREM

3.3.16: Let RG be a groupoid ring which is a E-ring. If RG has a proper

SNA-subring. A then RG is a S-E-ring.

Proof: Straightforward.

T

HEOREM

3.3.17: Let RG be a groupoid ring. Every S-E-ring need not in general be

an E-ring.

Proof: By an example. Consider the groupoid ring Z

G where G is given by the

following table:

. 1 a

1 e a

e a

P = {1, a

, a

, 1 + a

+ a

, 1 + a

+ a

, 1 + a

+ a

, a

+ a

} is a SNA-ring which

is a E-ring.

So RG is a Smarandache E-ring which is not a E-ring. Now we proceed on to define
pre-J-ring in groupoid rings.

D

EFINITION

3.3.17: Let RG be a groupoid ring. We say RG is a pre-J-ring if a

b =

for any pair a, b

∈ R G and n a positive integer.

EFINITION

3.3.18: Let RG be a groupoid ring we say RG is a Smarandache pre-J-

ring (S-pre-J-ring) if for every pair of elements a, b

∈ P, (where P is a proper subset

of RG which is a S-subring of RG) we have a

b =ab

for some positive integer n.

HEOREM

3.3.18: Suppose R is a pre-J-ring and G any groupoid, the groupoid ring

RG is a S-pre-J-ring but is general not a pre-J-ring.

Proof: Given R is a pre-J-ring, so for any groupoid, G we have RG to be a S-pre-J-
ring as R

⊂ RG. Now to prove RG in general is not a pre-J-ring even if RS is a S pre-

J-ring. Suppose G is a groupoid having elements like say a

b to exists and different

from ab

= 0 then we have the result to be true consider (1 + a

) = b and a = a

we see

(1 + a

) = 1+a

but a.b

= a

(1+a

) = 1 + a

but a.b

= a

, (1+ a

)

= 0. Hence the

claim.

Several such properties studied as in case of associative rings can also be found and
introduced in case of the special class of non-associative rings viz loop rings and
groupoid rings.

P

ROBLEMS

Does the groupoid ring Z

G where G = Z

(3,2) have pseudo commutative

pair 2?

2. Does the groupoid ring Z

G given in example 1 have a S-pseudo

commutative pair 2.

Find all the strongly regular elements of the groupoid ring Z

G where G =

(3, 5)

Does the groupoid ring given in problem 3 have S-strongly regular element?

Find a quasi commutative pair in Z

(3, 7).

Give an example of a groupoid ring which has no semi nilpotent elements.

Find groupoid rings, which has no semi nilpotent elements.

Give an example of a S-reduced groupoid ring.

9. Can

G where G = Z

(3,2) be a S-zero square ring? Justify your answer.

10.

Give an example of each of a groupoid ring which is a

a. E-ring.
b. S-E-ring.
c. p-ring.
d. S-p-ring.
e. pre-J-ring.
f. S-pre

J-ring.

g.  inner zero square ring.
h.  S-weak inner zero square ring.
i.  S-quasi commutative ring.

3.4. Smarandache substructures in Groupoid rings

This section is completely devoted to the introduction of Smarandache substructures
in groupoid rings. Study of groupoid rings is totally absent in literature, so the study
of Smarandache substructures in groupoid rings seems not only interesting but also
very new. These new class of non-associative rings is distinct from Lie algebras or
Jordan algebras or loop rings. First we introduce the concept of mod-p-envelope of a
groupoid ring. The notion of mod p envelope for group rings was first started [38] in
the year 1978, and later introduced in [55] in the year 1986 and [62]. For more about
group rings and above all finite loop rings refer [62].

D

EFINITION

3.4.1: Let G be a finite groupoid and Z

be the prime field of

characteristic p. Z

G be the groupoid ring of the groupoid G over Z

. The mod p-

envelope of G denoted by G

∗

= 1+U where

U =

{

}

∑

∈

∑

Example 3.4.1: Let Z

= {0, 1} be the prime field of characteristic two. G be the

groupoid given by the following table:

0 1 a

1 1 a

1 a

G is the groupoid ring of G over Z

. Now U = {0, 1 + a

1 ,

1 + a

, 1 + a

+ a

, a

+ a

, a

+ a

, …, a

+ a

, 1+ a

+ a

, …, a

+ a

}.G

∗

= 1 + U

= {1, a

, a

, 1+ a

+ a

, 1+ a

+ a

, …, 1 + a

+ a

}. It is easily

verified G

∗

is a groupoid under multiplication and G

⊂ G

∗

We have nice theorems about the mod p envelope of the groupoid G

∗

HEOREM

3.4.1: Let G be a commutative groupoid of order 2n in which square of

every element is one and Z

= {0,1} be the prime field of characteristic two. Then G

∗

is a commutative groupoid such that the square of every element is 1 and the order
of G

∗

is 2

2n-1

Proof: Let G be the groupoid with elements, G = {1, g

, g

, ….. g

2n-1

, g

= 1 and g

= g

, i = 1, 2, …, 2n – 1}. Z

G be the groupoid ring the mod p-envelop of G, G

∗

= 1

+ U where U = {0, 1 + g

, 1 + g

, … , 1 + g

2n – 1

, 1 + g

+ g

+ … + g

2n – 1

∗

= {1, g

, g

, …, g

2n-1

, …, g

+ g

+ … + g

2n-1

}. Now

α =

∈

∑α

is such that

= 1 as g

= 1 and g

, g

= g

and characteristic of Z

is two. Thus G

∗

is a

groupoid in which square of every element is 1 and G is strictly contained in G

∗

and

∗

| = 2

2n–1

. Clearly if G is non-commutative we may not be in a position to arrive at

this conclusion even if each

2
i

g = 1 in G.

Now we study the case when Z

is replaced by any other prime field Z

Example 3.4.2: Let Z

be the prime field of characteristic three. G be a groupoid

given by the following table:

∗

e a

e e a

e a

We see G is commutative with

2
i

a = e for each a

∈ G.

Now let Z

G be the groupoid ring with 1.e = e.1 = e=1 acting as the identity of Z

∗

= 1 + U = 1 + {0, 1 + 2a

, 1 + 2a

, 2 + a

, a

+ 2a

, a

+ 2a

, a

+ 2a

, a

+ 2a

, a

+ 2a

, 1 + a

+ a

, 1 + a

+ a

, 1+a

, a

+ a

, 2+

+ 2 a

, 2 + 2a

+ 2a

, 2 + 2a

+ 2a

, 2a

+ 2a

, 1+ a

+ 2a

1 +

+ 2a

, 1 + a

+ 2a

+2a

, 2 +

+ 2a

+ a

, 2+ a

+ a

+ 2a

, 2+2 a

+ a

} =

{1, 2 + 2a

, 2 + 2a

, a

, 2 + a

+ a

, 2 + a

+ a

, 2 + a

+ a

, 1 + a

+ a

, 2a

+ 2a

, 2a

+ 2a

, 2a

+ 2a

, 1 + 2a

+ 2a

, 2 + a

+ 2a

, 2 + a

+ 2a

, 2 + a

+ 2a

, 1 + a

+ 2a

, 1 + a

+ 2a

, 1 + a

+ 2a

, 1 + a

+ 2a

, a

+ a

, a

+ a

+ 2a

, 1 + a

+ 2a

, 1 + a

+ 2a

, 2a

+ a

}. Clearly G

∗

is closed

under the operation product and G

⊂ G

∗

. Further |G

∗

| = 27 = 3

= 3

4-1

We see only some of the elements are idempotents some of them units and so on.
In view of this we have the following theorem:

T

HEOREM

3.4.2: Let G be a commutative finite groupoid of even order with unit in

which every g

∈ G in such that g

= 1, Z

(p > 2) be the prime field of characteristic

p. Z

G be the groupoid ring of G over Z

. Then G

∗

= 1+ U the mod p envelope of G

is a groupoid of order p

2n-1

Proof: Given |G| = 2n and G is commutative with g

= 1 with unit. Z

G the groupoid

ring of G over Z

. G

∗

= 1+ U = 1+

}

{

∑

∈

∑

. It is easily verified

∗

is a groupoid of order p

2n-1

. Further G

⊂ G

∗

If p / n then the extra properties enjoyed by G

∗

is an interesting one.

Now we proceed on to define the Smarandache-mod-p-envelop of the groupoid G.

D

EFINITION

3.4.2: Let G be a finite groupoid. Z

the prime field of characteristic p.

G the groupoid ring of the groupoid G over Z

. The Smarandache mod p-envelope

(S-mod-p envelope) of G denoted by SG

∗

= 1 + U where U = {0,

Σα

∈ Z

B | g

∈ B

⊂ G; B a S-subgroupoid of G with Σα

= 0}.

Thus unlike the mod p-envelope of G we have for a given groupoid G several mod p-
envelopes. The importance is to find whether SG

∗

is a groupoid or a S-groupoid. If G

has no S-subgroupoid but G is a S-groupoid then we assume the mod-p-envelope of
G to be G

∗

= SG

∗

Thus the study is interesting only when the groupoids G has S-sub groupoids.

Example 3.4.3: Let Z

= {0, 1} be the prime field of characteristic two and G be the

groupoid given by the following table:

∗

e a

e e a

e a

Let Z

G be the groupoid ring. G is a S-groupoid, but G does not contain S-sub

groupoid. So G

∗

= SG

∗

Example 3.4.4: Let Z

= {0, 1} be the prime field of characteristic two and G be a

groupoid given by the following table.

∗

e a

e e a

G be the groupoid ring. Z

G has S-subgroupoids given by H = {e, a

, a

Identifying e with 1 we get. SG

∗

= 1 + {0, 1 + a

, 1 + a

, a

+ a

} = {1, a

, a

, 1 + a

}. Clearly H

⊆ SG

∗

, SG

∗

has unit for (1 + a

+ a

} = 1. H

= {a

, a

} is a S-sub

groupoid. Then SG

∗

= 1 + {a

+ a

, 0} = {1, 1 + a

+ a

}. Clearly H

⊆/ SG

∗

. Further

these S mod-p-envelopes are related.

Thus we have the following.

T

HEOREM

3.4.3: Let G be a groupoid with 1, Z

be the prime field of characteristic

two. The S-mod p envelope of G associated with proper S-sub groupoids H of G need
not in general have H

⊂ SG

∗

Proof: By an example. In the above example we have for H

= {a

, a

} which is a

S-subgroupoid of G. The SG

∗

associated with the S-subgroupoid H of G is such that

∗

= {1, 1 + a

+ a

}. Clearly H

⊆/ G

∗

. Hence the claim.

Now we proceed on to define the concept of Smarandache pseudo right ideals in
groupoid rings.

D

EFINITION

3.4.3: Let RG be a groupoid ring. B a proper subset of RG which is an

associative subring of RG. A non-empty subset X of RG is said to be a Smarandache
pseudo right ideal (S-pseudo right ideal) of RG related to B if

(1) (X, +) is an additive abelian group.
(2) For b

∈ B and s ∈ S we have sb ∈ S.

On similar lines we define Smarandache pseudo left ideal (S-pseudo left ideal). A
nonempty subset X of RG is said to be Smarandache pseudo ideal (S-pseudo ideal) of
RG if X is both a S-pseudo right ideal and S-pseudo left ideal.

Example 3.4.5: Let ZG be the groupoid ring of a groupoid G given by the following
table.

. e g

e e g

G is a S-groupoid, G has S-subgroupoid. Z

⊂ ZG is a S-groupoid ring for Z is the

associative subring. Let {e, g

, g

} = H. ZH

⊂ ZG and ZH is a S-pseudo ideal of

ZG.

D

EFINITION

3.4.4: Let RG be a groupoid ring which is a S-ring i.e. RG has a proper

subset A which is an associative ring. Let X be the S-pseudo ideal related to A. X is
said to be Smarandache minimal pseudo ideal (S-minimal pseudo ideal) of RG if X

is another S-pseudo ideal related to A and (0)

⊂ X

⊂ X implies X

= X or X

= {0}.

The minimality may vary with different related associative rings. On similar lines we
may define Smarandache maximal pseudo ideals (S-maximal pseudo ideals).

Let RG be a groupoid ring we say X is a Smarandache cyclic pseudo ideal (S-cyclic
pseudo ideal) related to the associative ring A if X is generated by a single element.
We say the S-pseudo ideal X of a groupoid ring RG related to the associative ring A,
A

⊂ RG is Smarandache prime pseudo ideal (S-prime pseudo ideal) related to A if xy

∈ I implies x ∈ I or y ∈ I.

The study of groupoid rings in mathematical literature is completely absent so the
study of these ideals S-ideals, pseudo ideals and S-pseudo ideals will be very
innovative and interesting.

The first open problem is, will the set of S-ideals of a finite groupoid ring of any
groupoid G over a ring R form a modular lattice. Atleast study groupoids in Z

∗∗∗

(G)

over Z

. Z

may be a prime field of characteristic p or a ring with p-elements that is p

is a composite number.

The study of A.C.C. and D.C.C conditions on groupoid rings is yet another
interesting feature for in case of group ring such study have been made. Several
unsolved problems exist even in case of group rings to be Noetherian. It has been
proved that KG is Artinian if and only if G is finite. Is it true, KG is Noetherian if
and only if G has a series of subgroups?

G = G

⊃ G

⊃ … ⊃ G

= {1} such that G

i+1

is normal in G

and G

i+1

is either a

finite group or infinite cyclic.

Now the analogous study for groupoid rings remains untouched. Analogous to group
rings which satisfy S-ACC or S-D CC we can define for the groupoid rings RG to
satisfy S-ACC and S-D.C.C on S-ideals.

EFINITION

3.4.5: Let RG be a groupoid ring. If the set of all S-right ideals of RG is

totally ordered by inclusion we say RG is Smarandache right chain groupoid ring (S-
right chain groupoid ring).

D

EFINITION

3.4.6: Let RG be a groupoid ring, A a S-subring of RG. If the set of all

S-ideals of A in RG is totally ordered by inclusion then the groupoid ring RG is said
to be a Smarandache weakly chain groupoid ring (S-weakly chain groupoid ring).

D

EFINITION

3.4.7: Let RG be a groupoid ring. I

≠ 0 be an ideal of RG . If for any

nontrivial ideals X and Y of RG, X

≠ Y we have 〈 X ∩ I, Y ∩ I〉 = 〈 X, Y〉 ∩ I then I is

called the obedient ideal of RG,

〈〉 denotes the ideal generated by the collection of

elements in X

∩ I and Y ∩ I.

EFINITION

3.4.8: Let RG be a groupoid ring. If for any S-ideal I of RG we can

find obedient S-ideals of RG say X, Y in RG, X

≠ Y such that 〈 X ∩ I, Y ∩ I〉 = 〈 X, Y〉

∩ I. Then we say I is a Smarandache obedient ideal of RG (S-obedient ideal). If in a
groupoid ring RG we have every S-ideal I of RG to be a S-obedient ideal of RG then
we say RG is a Smarandache ideally obedient ring (S-ideally obedient ring).

Now we give a nice result about groupoid rings, which are not true in case loop rings
or group rings.

T

HEOREM

3.4.4: Let RG be a groupoid ring. If G has no unit and no subgroupoids

and R is a S-ideally obedient ring, then RG is an S-ideally obedient ring.

Proof: We know every I

≠ 0 in R is a S-obedient ideal, so is every I G ≠ 0 in RG as

G has no sub groupoids. Hence the theorem.

Example 3.4.6: Let Z

= {0, 1, 2, …, 11} be the ring of integers modulo 12. Let G

be a groupoid given by the following table:

∗

Clearly the groupoid ring Z

G has S-obedient ideals as Z

has I={0,6} to be an

obedient ideal of Z

. When we try to find non-associative Lin rings we will not be in

a position to say nontrivially all Lie algebras are Lin rings as (xy – yx)

= xy – yx or

(xy + yx)

= xy + yx.

EFINITION

3.4.9: Let RG be a groupoid ring. We say RG is a Lin ring if (xy – yx)

= xy – yx or (xy + yx)

= xy + yx for every x , y in R and for some n = n (x , y) > 1.

HEOREM

3.4.5: Let Z

G be a groupoid ring where Z

= {0,1} and G a commutative

groupoid with unit in which every g

∈ G is such that g

= 1 is a Lin Ring.

Proof: Obvious by the very definition of Lin ring.

D

EFINITION

3.4.10: Let RG be a groupoid ring. We say RG is an ideally strong ring

if every subring of RG not containing identity is an ideal of RG. We call RG a
Smarandache ideally strong ring (S-ideally strong ring) if every S-subring of RG is
an S-ideal of RG.

We define I

∗

rings in case of groupoid rings.

EFINITION

3.4.11: Let RG be a groupoid ring. {I

} be the collection of all ideals of

RG. RG is said to be a I

∗

ring if every pair of ideals I

, I

∈ {I

} in RG and for every

∈ RG \ (I

∪ I

) we have

〈a ∪ I

〉 = 〈 a ∪ I

〉 where 〈〉 denotes the ideal

generated by a and I

; j = 1, 2.

Now we define Smarandache I

∗

ring (S-I

∗

-ring) if for all S-ideals {A

} of RG; we have

for every pair of ideals A

, A

∈ {A

}; and for every x

∈ RG \ {A

∪ A

〈A

∪ x 〉 =

〈A

∪ x 〉 and they generate S-ideals of RG. We call the ring to be a Smarandache

weakly I

∗

ring (S-weakly I

∗

-ring) if for some x

∈ R \ {A

∪ A

} we have

〈A

∪ x〉 =

〈A

∪ x〉.

HEOREM

3.4.6: All S-I

∗

rings are S-weakly I

∗

rings.

Proof: Follows from the very definition.

The concept of Q-rings in case of groupoid rings and all the more in case of non-
associative rings remains practically absent. So now we introduce Q-ring only in
case of groupoid rings.

D

EFINITION

3.4.12: Let RG and R

be any two non-isomorphic finite groupoid

rings, if there exists non-maximal ideals I if RG and J of R

such that RG/I is

isomorphic to R

/ J

Then the finite groupoid rings are said to be Q-groupoid

rings.

Example 3.4.7: Z

G and Z

be two finite groupoid rings where G

∈ Z

(2,3) and

∈ Z

(1,3). Can Z

G and Z

be Q-groupoid rings. We define further the concept

of weakly Q-ring.

D

EFINITION

3.4.13: Let RG be a groupoid ring such that all of its ideals are

maximal and if we have RG/{0} is isomorphic to some groupoid ring R

; then we

call RG a weakly Q-ring.

D

EFINITION

3.4.14: Let RG be a groupoid ring and A, a S-ideal of RG. RG/A is

defined as the Smarandache quotient ring (S-quotient ring) related to the S-ideal A
of RG.

EFINITION

3.4.15: Let RG and R

be any two groupoid rings, if we have S-ideals

A and B of RG and R

respectively such that the S quotient ring RG/A is S-

isomorphic with the S-quotient ring R

/B then we say RG is a Smarandache Q-

ring (S-Q-ring).

Several examples and results can be got using the groupoids from the new class of
groupoids Z

∗∗∗

(n) and using the modulo ring of integers Z

EFINITION

3.4.16: Let RG be a groupoid ring. RG is called an F-ring if there

exists a finite set X of non-zero elements in RG such that a RG

∩ X ≠ φ for any non-

zero a in RG.

If in addition X is contained in the center of RG, RG is called a FZ-ring.

For the groupoid ring RG we say a S-subring A of RG is a Smarandache F-ring (S-
F-ring) if we have a subset X in RG and a non-zero b

∈ RG such that bA ∩ X ≠ φ.

Here it is pertinent to mention that we need not take X as a subset of A but nothing is
lost even if we take X to be a subset of A. Similarly b can be in A or in RG. In case of
associative rings one defines the concept of

-rings.

EFINITION

3.4.17: Let RG be a groupoid ring. RG is said to be a

-ring if for n >1,

n an integer x

– x is an idempotent for all x

∈ R.

EFINITION

3.4.18: Let RG be a groupoid ring. RG is a Smarandache

-ring (S-

ring) if for every x

∈ RG, x

– x is a S-idempotent for some integer n >1.

HEOREM

3.4.7: If RG is a Smarandache

-ring then RG is a

-ring.

Proof: By the very definitions the result follows.

T

HEOREM

3.4.8: Let Z

G be the groupoid ring where G is a commutative groupoid

with unit in which every g

∈ G is such that g

= 1. Then Z

G is not a

-ring.

Proof: Follows from the fact that RG has elements

α ≠ 0 such that α

=0 so

can never be an idempotent. Hence the claim.

D

EFINITION

3.4.19: Let RG be a groupoid ring. A non-empty subset S of RG is called

a closed net of RG if S is a closed set of RG under ‘.’ and is generated by a single
element. That is S is a groupoid.

If the groupoid ring RG is contained in a finite union of closed nets of RG then we
say RG has a closed net.

If RG is a groupoid ring RG is a CN-ring if RG =

∪ S

where S

‘s are closed nets

such that S

∩ S

’s are closed nets such that S

∩ S

φ or {1} or {0} . if i ≠ j and 1

∈ RG and S

∩ S

= S

if i = j and each S

is a nontrivial closed net of R.

RG is said to be a Smarandache closed net if

1. S is a groupoid
2. S is a S-groupoid

ROBLEMS

Find all SG

∗

for G = Z

(3,4) over the prime field Z

2. Find

∗

for G = Z

(2,5) over the prime field Z

. Is SG

∗

= G

∗

Whether there exists a relation between G

∗

and atleast one of the SG

∗

Let G = Z

(5,7) be a groupoid. Does ZG have S-pseudo ideals?

Give an example of a groupoid ring, which has no S-pseudo ideals.

Does the groupoid ring ZG where G = Z

(5,7) satisfy A.C.C. or D.C.C or

S.A.C.C or S.D.C.C.

Give an example of a groupoid ring which satifies A.C.C and not S.A.C.C.

Give an example of a groupoid ring which satisfies both S.D.C.C and D.C.C
on S-ideals.

Find the set of all S-ideals of the groupoid ring Z

G where G = Z

(5,12).

10.

Give an example of a groupoid ring, which is a

-ring.

11.

Is the groupoid ring Z

G where G = Z

(5, 3) a

- ring? Justify your answer.

12.

Give an example of a groupoid ring, which is a F-ring.

13.

Can the groupoid ring Z

G where G = Z

(2,3) be a F-ring? Justify your claim.

14.

Give an example of a groupoid ring which is a CN-ring.

15.

Is the groupoid ring Z

G where G = Z

(3,6) a CN –ring?

16.

Can the groupoid ring given in example (15) be a S-CN –ring?

17.

Does the groupoid ring Z

G where G = Z

(4,8) have a S-obedient ideal?

18. Will the groupoid ring Z

G where G = Z

(3,9) have a obedient ideal?

3.5 Special Properties in Groupoid rings and Smarandache Groupoid
rings

This section is solely devoted to the introduction of several interesting properties in
groupoid rings.  The main concepts introduced in this section are the n-capacitor sub
groups, essential and S-essential groupoid rings, exponential groupoid rings and
Smarandache exponential rings and groupoid rings satisfying special identity (xy) x
= x(yx). Finally we introduce the concept of subring link relation in groupoid rings.
Radicals and S-radicals are introduced and studied in case of groupoid ring RG.
These studies are carried out both in case of finite and infinite groupoid rings.

Here we introduce the concept of a new class of groupoids using [-

∝, ∝].

EFINITION

3.5.1: Let L = [-

∝, ∝]. Define for a , b ∈ L, a ∗ b = ma + nb where m,

∈ L. Clearly {L, ∗, (m, n)} is a groupoid of infinite order. For varying m, n in L we

get an infinite class of infinite groupoids.

Result 1: [L,

∗, (1, 1)] is a semigroup.

Result 2: The groupoids [L,

∗, (m, n)] ≠ [L, ∗, (n, m)] when m ≠ n.

Result 3: The groupoid [L,

∗, (m, n)] satisfies the identity (x ∗ y) ∗ (z ∗ y) = x ∗ z if

and only if m = 1 and n = 0 or m = 1 and n = -1.

In view of these results we can study the groupoid ring RL where L is a groupoid
built using [-

∝, ∝].

HEOREM

3.5.1: Let L = {[-

∝, ∝], ∗, (m, n)  m, n ∈ [-∝, ∝]} be an infinite

groupoid. Z

= {0, 1} be the prime field of characteristic two. Z

L is an infinite non-

commutative groupoid ring if m

≠ n of characteristic two and has every element to be

of order 2 if m = -n.

Proof: Straightforward by the very definition.

It is left for the author to built interesting and innovative results in this direction by
incorporating the properties of Smarandache substructures introduced in section 4
and Smarandache special elements in section 3.

Remark: Using these groupoids L and choosing any Z

(p <

∝), p a prime. Z

L will

be an infinite groupoid ring of characteristic p. If both m and n are chosen to be
positive we see every element in L is torsion free and the groupoid L will be a
torsion free non-abelian groupoid.

D

EFINITION

3.5.2: Let RG be a groupoid ring. We say a

∈ RG (a ≠ 0) is said to be a

normal element of RG if a RG = RG a. If every element a

∈ R G is a normal element

then we call RG a normal groupoid ring.

We give nice condition for a groupoid ring to be a normal groupoid ring.

T

HEOREM

3.5.2: Let K be any field and G a groupoid such that G is a normal

groupoid then the groupoid ring KG is a normal groupoid ring.

Proof: Straightforward by the very definition.

D

EFINITION

3.5.3: Let KG be a groupoid ring We say KG is a Smarandache normal

groupoid (S-normal groupoid) ring if KG has a proper subset A, A

⊂ KG and A a

SNA-subring of KG, such that for all

α ∈ K G, αA = Aα.

Thus we have the immediate consequence.

T

HEOREM

3.5.3: Suppose KG is a S-normal groupoid then KG need not in general

be a normal groupoid.

Proof: The reader is requested to construct an example to prove this result.

The natural question would be if KG is a normal groupoid will KG be a S-normal
groupoid, even this question is not that easy to be settled.

HEOREM

3.5.4: Let G be a groupoid. K any field, KG the groupoid ring. g

∈ G is a

normal element of KG if and only if gG = Gg.

Proof: One way is obvious if gG = Gg then gKG = KGg .

Suppose g

∈ G is such that gKG = KGg then we must have gG = Gg for otherwise

we will not have gKG = KGg as G

⊂ KG. It is interesting to note all elements of G

are not normal elements of KG for if gG

≠ Gg then gKG ≠ KGg. This is clear by the

example. For take, Z

= {0,1} and G a groupoid given by the following table:

. a

Clearly in the groupoid ring Z

G we have a

≠ Z

G a

as a

≠ Ga

. Hence the

claim.

D

EFINITION

3.5.4: Let KG be a groupoid ring N(KG) = { set of all elements in KG

such that

αKG = KGα where α ∈ KG} that is N(KG) is the set of normal elements

in KG.

T

HEOREM

3.5.5: Let KG be a groupoid ring. N(KG) is not even closed under,'.' .

Proof: By an example. Let K be any field and G be a groupoid given by the
following table:

∗ a b c
a b c a
b a b c
c c c c

aKG = KGa for a

∈ G ⊂ KG but a.b = b ∉ N (KG) as b.KG ≠ KGb. Hence the

claim.

Now we study the condition for a groupoid ring RG to satisfy the identify (

α β) α =

α (βα) for all α, β ∈ RG.

Example 3.5.1: Consider the groupoid ring Z

G where G is given by the following

table:

. a

It is easily verified, the groupoid ring Z

G satisfies the identify (

α β)α = α(αβ) for

all

α, β ∈ Z

Example 3.5.2: Let Z

G be a groupoid ring where G is a groupoid given by the

following table:

. a

Clearly the groupoid ring does not satisfy the identity for (a

) a

≠ a

) for a

∈ G; so for all α, β ∈ Z

G we see (

αβ) α ≠ (βα). Hence all groupoid rings in

general do not satisfy the identity.

[46] has introduced the concept of exponentiation in a ring R. He says the triple (R,
B, E) is an exponential ring if for the ring R, B is the multiplicative subsemigroup of
(R, .) (B the basis) which does not contain zero and E (the exponents) be a semiring
with unit element 1; that is (E, +) is a semigroup and (E,.) is a semigroup with 1 and
‘.’ distributes over ‘+’ both from the left and from the right with a binary operation
from

B

× E → B ⊆ R (that is (b, e) → b

) such that for all b, d

∈ B and e, k ∈ E , b

(bd)

. b

= (b

)

, b

e+k

= b

and b' = b.

Using this concept we build non-associative rings by replacing the associative rings
which we call as the first type of exponentiation .The second type of exponentiation
would be by replacing in the triple (R, B, E); E the exponents by a groupoid under +
and a groupoid under ‘.’ with 1 and + and ‘.’ distributors over each other.

For more about exponentiation please refer [46]

D

EFINITION

3.5.5: Let A = RG be a groupoid ring of a groupoid G with 1. Let P be a

multiplicative groupoid of (RG,.) which does not contain zero and let E (the
exponents) be a semi ring with unit 1.

A binary operation ‘

×’ from P × E → P ⊆ A = RG is ((p,e) → p

) makes (RG, P, E) a

non-associative ring if it has the usual properties of exponentiation viz. for all b, d

∈

P and e, k

∈ E , b

=(bd)

= (b

)

, b

e+k

= b

and b' = b. Thus the

exponentiation in a non-associative ring is just defined from the exponentiation of a
ring by replacing the semigroup by a groupoid.

Example 3.5.3: Let R be a commutative ring with 1 or a field. L a loop so that Lis
trivially a groupoid .RL be the loop ring. Take P a multiplicative groupoid from RL

where P does not contain 0. Take E exponents and (RL, P, E) is the exponential ring
of the non-associative ring RL.

Clearly (R, S, E) where R is a ring and S is a subsemigroup of (R, .) without zero is
the exponential subring of (RL, P, E). Thus we can build exponential non-associative
rings from exponential associative rings. If we replace the loop by a groupoid with 1
still one gets a non-associative ring on which exponentiation can be defined.

D

EFINITION

3.5.6: Let (RG, P, E) be a non-associative exponentiation ring. We say

(RG, P, E) is a Smarandache non-associative exponentiation ring (S-non-associative
exponentiation ring) if this has a sub triple (R, S, E), which is an associative ring
with exponentiation.

Thus we have got the following theorem the proof of which is straightforward.

T

HEOREM

3.5.6: Let RG be a groupoid ring (where G is a groupoid with 1) or a

loop ring. Then the triple (RG, P, E) is a S-non-associative exponentiation ring.

We define n-capacitor for commutative groupoid rings.

D

EFINITION

3.5.7: Let RG be a commutative groupoid ring P an additive subgroup

of RG, P is called the n-capacitor group of RG if x

⊂ Pfor every x ∈ R and n >1

and n a positive integer.

Example 3.5.4: Let Z

G be a commutative groupoid ring where the groupoid G is

given by the following table:

∗

G is a commutative groupoid ring. Let P = Z

, a

}. It is easily verified P is

a n capacitor group of Z

EFINITION

3.5.8: Let RG be a commutative groupoid ring and P an additive S-

semigroup of RG. P is called a Smarandache n-capacitor group (S-n-capacitor
group) of RG if x

⊆ P for every x ∈ RG and n ≥ 1; n a positive integer.

We now proceed on to define radical for groupoid rings.

D

EFINITION

3.5.9: Let RG be a groupoid ring. A subset P of RG is a radical if

(1). P is an ideal of RG
(2). P is a nil ideal
(3). R/P has no non-zero nilpotent right ideals.

The sum of all ideals in RG satisfying (1) and (2) is the upper radical of RG and is
denoted by U (RG). The intersection of all these ideals in RG satisfying (1) and (3)
is the lower radical of RG denoted L(RG).

D

EFINITION

3.5.10: Let RG be a groupoid ring. The Smarandache radical ideal (S-

radical ideal) P of RG is defined as follows:

1. P is a S-ideal of RG
2. S

⊂ P where S is a subideal of P is a nil ideal

3. RG/P has no non-zero nilpotent right ideals.

The sum of all S-ideals of RG satisfying (1) and (2) is called the Smarandache upper
radical (S-upper radical) of RG and is denoted by S(U (RG)). The intersection of
those S-ideals in RG satisfying (1) and (3) is the Smarandache lower radical (S-
lower radical) of RG denoted by S(L(RG)).

Now we proceed on to define the subring link relation in groupoid rings.

D

EFINITION

3.5.11: Let RG be a groupoid ring. A pair of elements x,y

∈ RG is said

to have a subring right link relation if there exists a subring M of RG in RG\ {x, y}
that is M

⊆ RG \ {x, y} such that x ∈ My and y ∈ Mx. Similarly subring left link

relation if x

∈ y M and y ∈ xM. If it has both a left and a right link relation for the

same subring M then we have x and y to have a subring link relation and is denoted
by xMy.

D

EFINITION

3.5.12: Let RG be a groupoid ring. We say a pair x, y in RG has a

weakly subring link with a subring P in RG\ {x , y} if either y

∈ Px or x ∈ Py ‘or’ in

the strictly mutually exclusive sense and we have subring Q, Q

≠ P such that y ∈ Q x

(or x

∈ Q y).

We say a pair x, y in RG is one way weakly subring link related if we have a subring
P

⊂ RG \ {x, y} such that x ∈ Py and for no subring S ⊂ RG \ {x,y} we have y ∈ Sx.

Let RG be a groupoid ring a pair x ,y

∈ R G is said to have a Smarandache subring

link relation (S-subring left link relation) if there exists a S-subring P in RG\{x , y}
such that x

∈ Py and y ∈ xP. If it has both a Smarandache left and right link

relation for the same S-subring P then we say x and y have a Smarandache subring
link (S-subring link).

We say x, y

∈ RG is a Smarandache weak subring link (S-weak subring link) with a

S-subring P in RG \ {x, y} if either x

∈ Py or y ∈ Px (or in strictly mutually exclusive

sense) we have a S-subring Q

≠ P such that y ∈ Q x (or x ∈ Qy).

We say pair x, y

∈ RG is said to be Smarandache one way weakly subring link

related (S-one way weakly link related) if we have a S-subring P

⊂ RG \{x, y} such

that x

∈ Py and for no subring Q ⊆ RG \{x , y} we have y ∈ Qx.

Let us now define essential subring of the groupoid ring RG.

EFINITION

3.5.13: Let RG be a groupoid ring; A S-subring x of RG is said to be a

Smarandache essential subring (S-essential subring) of RG if the intersection of X
with every other S-subring is zero. If every S-subring of RG is an S-essential subring
of RG then we call RG an S-essential ring.

Study of essential subrings, S-essential subrings and groupoid rings, which are
essential rings, happens to be an interesting field.

Example 3.5.5: RG = Z

G be the groupoid ring where G is the groupoid given by the

following table:

. a

The groupoid ring Z

G has Z

], Z

, a

] and Z

] to be

subgroupoid rings. We see Z

G is not an essential ring or does not even have a single

essential subring.

P

ROBLEMS

Construct an infinite groupoid ring of characteristic two using Z

= {1, 2, 3,

…}.

Give an example of a normal groupoid ring.

Find an example of a S-normal groupoid ring.

Characterize a class of groupoid rings RG which satisfy the identify (

αβ)α =

α(βα) for α, β ∈ RG .

5. Is the groupoid ring Z

G be made into an exponential ring where G

∈

∗∗∗(9).

Construct rings, which are not S-exponential rings.

Characterize those groupoid rings, RG which are n-capacitor rings G taken
from Z

∗∗∗(p) and R = Z

Find the radical and S- radical for the groupoid rings Z

G where G

∈ Z∗∗∗

(n) where

a. (m, n) = 1.
b. (m, n) = d

≠ 1 or n

c. (m, n) = n.
d. m = n.

Classify those groupoid rings R G which are

a. essential or
b. S-essential.

when G

∈ Z∗∗(7) and R = Z

10.

Characterize those groupoid rings RG which are never essential or S-essential
when R= Z

and G

∈ Z∗(6).

Chapter 4

LIE ALGEBRAS AND SMARANDACHE LIE
ALGEBRAS

In this chapter we recall the definition of Lie algebras and its properties and
introduce the concept of Smarandache Lie algebras. This Chapter has five sections.
In the first section we introduce the basic notions of Lie algebras. Section two
introduces Smarandache Lie algebras and basic properties of Lie algebras. Section
three is devoted to defining several new notions in Smarandache Lie algebras and
Lie algebras. Section four is devoted to the study of certain special properties and in
the final section we introduce some new concepts in Lie algebras and S-Lie algebras
and define Smarandache mixed direct product of Lie algebras of type A and B. As
every section has an introduction we do not recall the contents of each section here.

4.1 Basic Properties of Lie algebras

In this section we just recall the definition of real and complex Lie algebras from
[36, 74] and some of the basic properties enjoyed by them: but we do not claim that
we have fully exhausted recalling all the basic properties of the Lie algebras. We
have enlisted only some of the properties, which has interested us. We leave it for the
reader to refer books on Lie algebras for more literature.

D

EFINITION

4.1.1: A real Lie algebra L is a vector space over the real field

furnished with a bilinear product [

λ, µ] satisfying the identities.

[

λ, µ] = - [λ, µ]

and [ [

λ, µ], ν] + [ [µ ,ν],λ]+ [[ν, λ],µ]=0

[Jacohi identity]

In what follows we will always suppose unless the contrary is explicitly specified,
that a Lie algebra is finite dimensional. Now we just recall the definition of complex
Lie algebra.

D

EFINITION

4.1.2: Let L be a real Lie algebra. Let L be the set of all formal sums

+ i

µ; λ, µ ∈ L. We define addition and multiplication by complex numbers and a

product [

λ+ iµ, λ' + iµ'] by the formula (λ + iµ) + (λ' + iµ') = λ + λ ' + i(µ + µ')

(x +iy) (

λ + iµ) = (xλ -yµ) + i(yλ + xµ), [λ + iµ , λ' + iµ'] = ([λ'λ'] - [µ' µ']) + i

([

λ, µ'] + [µ, λ']).

It is easily verified that with these definition L becomes a complex Lie algebra. The
complex Lie algebra L is called the complexification of the real Lie algebra L.

Thus we can also say non-associative algebra L is said to be associative if its
multiplication satisfies the associative law that is (xy) z = x (yz).

A non-associative algebra L is a Lie algebra, if its multiplication satisfies the Lie
conditions i.e., x

= 0, (xy) z + (yz)x + (zx)y = 0, that is the Jacobi identify is true.

Clearly if L is a Lie algebra and x, y

∈ L then 0 = (x + y)

= x

+ xy + yx + y

= 0 so

that xy = - yx. Conversely if this condition holds 2x

= 0 so that if the characteristic

of L is not two then x

= 0. Thus for characteristic not two we can assume instead of

= 0; xy + yx = 0 or xy = - yx. In general if two elements x,y

∈ L is such that [x, y]

= 0 then we say x and y commute . The Lie algebra L is abelian if [ x , y ] = 0 for all
x , y

∈ L.

Example 4.1.1: Let C

be a three dimensional vector space over C and e

, e

be a

basis of C. For any two elements x = x

+ x

, and y = y

+ y

define [x, y] = (x

– x

) e

+ (x

– y

) e

+ (x

– x

. Then C

becomes a

Lie algebra.

Example 4.1.2: Let S denote the collection of all 3

× 3 matrices with entries from C.

For any X, Y

∈ S define [X, Y] = XY – YX, then S is a Lie algebra.

EFINITION

4.1.3: Let L and L

be any two Lie algebras. A one to one map from L

onto L

is called a Lie algebra isomorphism if it satisfies:

(1) if X

→ Y

, X

→ Y

then for any

λ , µ ∈ C (where L and L

are Lie algebras

over C)

λX

µX

→ λY

µY

(2) if X

→ Y

, X

→ Y

then

, X

]

→ [Y

, Y

We say L and L

are isomorphic and denote it by L

≅ L

. In particular an

isomorphism from L onto itself is called an automorphism.

One of the basic fundamental problems in Lie algebras is to determine all non-
isomorphic Lie algebras. The following is a very important and commonly used
example.

Example. 4.1.3: Let gl (n, C) be the set of all n

× n matrices. It is well known with

respect to matrix addition and scalar multiplication gl(n, C) forms n

dimensional

vector space.

Now for any X , Y

∈ gl(n, C), define [X, Y] = XY – YX, then gl (n, C) forms a Lie

algebra. gl(n, C) can be considered as the set of all linear transformation of an n-
dimensional vector space V, then in this case it is denoted by gl

(V).

EFINITION

4.1.4: Let L be a Lie algebra. M and N denote subsets of L. Denote by

M + N the linear subspace spanned by elements of the form m + n (m

∈ m, n ∈ N )

and by [M, N] the subspace spanned by elements of the form [m , n] (m

∈ M and n ∈

N) . If M, M

, M

, N

P are subspaces of L then

(1) [M

+ M

, N]

⊆ [M

, N] + [M

, N].

(2) [M, N] = [N, M].

(3) [M, [N , P]]

⊆ [N, [P, M] ] + [P, [ M,N]].

A subspace H of L is said to be a subalgebra if [H,H]

⊆ H, i.e. X, Y ∈ H implies that

[X , Y]

∈ H. A subspace H of L is said to be an ideal if [L , H] ⊆ L i.e. X ∈ L and Y

∈ H implies [X , Y ] ∈ H.

An ideal is a subalgebra if H

and H

are ideals then H

+ H

and H

∩ H

are also

ideals of L. Subalgebras of gl(n, C) are called matrix Lie algebras or linear Lie
algebras. If H is an ideal of L, then the quotient L/H which consists of all cosets
(congruence classes mod H) is defined. For X

∈ L denote the congruence class

containing X by X . Define [X, Y] =

]

[

the definition is independent of X and

Y.

The quotient space L/H thus becomes a Lie algebra, this algebra is called the
quotient algebra of L with respect to H. If L is a Lie algebra and H be an ideal of L,
then it can be proved that the mapping X

→ X satisfies the conditions.

(1) if X

→ X , Y → Y then for any λ , µ ∈ C, λ X + µ X → λ X + µ Y .

(2) if X

→ X , Y → Y then [X , Y ] → [ X , Y ].

In general a mapping X

→ X

from a Lie algebra L to Lie algebra L

is said to be a

homomorphism if it satisfies;

(1) if X

→ X

, Y

→ Y

then for any

λ , µ ∈ C, λ X + µ Y → λ X

µ Y

(2) if X

→ X

, Y

→ Y

then [X , Y]

→ [X

, Y

If this mapping is onto then L

is said to be a homomorphism image of L.

Example 4.1.4: All trace zero matrices of gl(n, C) form a subalgebra, denote it by
A

n-1

. In fact A

n-1

is an ideal of gl(n, C) for if X, Y

∈ gl(n, C) then Tr [X, Y] = Tr [XY

– YX ] = 0, therefore [X, Y]

∈ A

n-1

. All scalar matrices of gl(n, C) form a one

dimensional subalgebra which is also an ideal of gl(n, C) for if

λI is a scalar matrix

then for any x

∈ gl(n, C) we have [X , λ I] = XλI – λIX = 0. All diagonal matrices of

gl(n, C) form an n-dimensional abelian subalgebra; denote it by d (n, C). The set of
trace zero diagonal matrices form an (n – 1) dimensional abelian subalgebra of A

n-1

EFINITION

4.1.5: Let L be a Lie algebra. Obviously L and {0} are ideals of L. If L

does not have any other ideals, then it is said to be a simple Lie algebra.

EFINITION

4.1.6: Let L be a Lie algebra and M

, …, M

be ideals of L, if any

element X in L can be uniquely written as X = X

+ X

+…+ X

where X

∈ L, i = 1,

2,…, n, then L is said to be the direct sum of M

, …, M

. We also denote direct sum

by L = M

+ …+ M

. If L is a direct sum of M

, M

, …, M

then for i

≠ j, M

∩ M

{0}, for if X

∈ M

∩ M

, then X = 0 +…+ X

+…+ 0 = 0 + …+ X

+…+0 are two

expressions of X and this forces X = 0. From M

∩ M

= {0} (i

≠ j), it follows that

[Mi, Mj] = {0} (i

≠ j) are ideals.

EFINITION

4.1.7: Let L be a Lie algebra, [L, L] is called the derived algebra of L

and is denoted by DL. If H is an ideal of L then so is DH. In fact [L, DH] = [L, [H,
H]]

⊆ [H, [H, L]] + [H, [L, H]] ⊆ [H, H] + [H, H ] = DH. D

L = L, D

L = DL,

n+1

L = D(D

L). In this way, we obtain a series of ideals satisfying D

⊇ D

⊇

…

⊃ D

⊃ …

This series is called the derived series of L. If there exists a positive integer n such
that D

L = {0} then L is called a solvable Lie algebra. The following are true about

solvable Lie algebras.

Subalgebras of solvable algebras are solvable, homomorphic images of solvable
algebras are solvable, quotient algebras of solvable algebras are solvable. If L is a
Lie algebras H an ideal and if H and L / H are solvable then L is solvable. Direct
sum of solvable Lie algebras are solvable.

If an ideal H of a Lie algebra L is solvable then it is called a solvable ideal.

If L does not contain any solvable ideal except {0} then L is said to be semi simple.

An equivalent definition is that L is semi simple if it does not contain any non-zero
abelian ideal. In fact non-zero abelian ideals are certainly solvable.

A semi simple Lie algebra is the direct sum of all the minimal ideals.

The one-dimensional Lie algebra is the only Lie algebra which is simple but not semi
simple. So in this text the one-dimensional Lie algebra will be excluded from the
simple Lie algebras, hence all simple Lie algebras are semi simple.

If L has a unique maximal solvable ideal

τ , τ is called the radical of L.

EFINITION

4.1.8: A Lie algebra L is said to be nilpotent if there exists a positive

integer n such that C

L = {0}, where C

L = L, C

L = [L, C

L ], …, C

(n-1)

L = [L,

L] = … If C

n-1

L is an ideal of L then [L, C

⊆ [L, C

(n-1)

L] = C

L, hence C

L is

also an ideal of L. In this way we obtain a series of subalgebras satisfying C

⊇

⊇ … ⊇ C

(n)

⊇ … and each C

(n)

L is an ideal of L. This series is called the

descending central series. If there exists a positive integer n such that C

(n)

L = {0},

then L is said to be nilpotent, that is L is nilpotent for some positive integer n, [X

[… [X

] …]] = 0 for any X

, …, X

∈ L. It can be proved D

(n)

⊆ C

(n)

L. In fact

from D

⊆ C

L, it follows that D

(n+1)

L = [ D

(n)

L , D

(n)

⊆ [L , C

(n)

L] = C

(n+1)

Some of the interesting properties on nilpotent Lie algebras are:

1. If L is nilpotent then it is solvable.

2. Subalgebras of nilpotent Lie algebras are nilpotent. Homomorphic

images of nilpotent Lie algebras are nilpotent in particular quotient
algebras of nilpotent Lie algebras are nilpotent.

Direct sums of nilpotent algebras are nilpotent.

Example 4.1.5: If t (n, C) denote the set of all upper triangular matrices of gl(n, C)
i.e. matrices of the form













is a Lie algebra t(n, C) . Elements in t(n, C) that have equal diagonal entries form a
Lie algebra m(n, C); t(n, C) is solvable and m(n, C) is nilpotent.

The image of the mapping A

→ adA is denoted by adL; the kernel of this mapping is

the set of all A

∈ L satisfying [A, X] = 0 for all X ∈ L. These elements are called

central elements.

For a linear transformation A in a finite dimensional vector space V the fitting
lemma asserts that V = V

⊕ V

where V

are invariant relative to A and the

induced transformation in V

is nilpotent and in V

it is an isomorphism , [36] calls

A the fitting null component of V relative to A and V

fitting one component of V

relative to A. The space V

= {x / xA

= 0 for some i} and

∞

∩

Let L be a Lie algebra of linear transformations in a finite dimensional vector space
over a field of characteristic 0. An element h

∈ L is called regular if the

dimensionality of the Fitting null component of L relative to adh is minimal. A Lie
algebra L is a linear transformations of a finite dimensional vector space over a field
of characteristic zero is called almost algebraic if it contains the nilpotent and semi
simple components for every X

∈ L.

For more about Lie algebras refer [36, 74].

4.2 Smarandache Lie Algebras and its Basic Properties

In this section we introduce the concept of Smarandache Lie algebras Till date
Smarandache Lie algebras were not defined. To define this concept we basically
need the notion of Smarandache vector space, Smarandache linear transformation
etc, but except for the definition of Smarandache vector space nothing has been
developed about Smarandache linear transformation etc.

So in chapter one some basic definitions about Smarandache vector spaces have been
introduced mainly keeping in mind the essential concepts used in the definition of
Smarandache Lie algebras. Further we do not even try to define Lie groups for we
felt the very study of Smarandache Lie groups can be taken as a separate one, hence
we have defined only Smarandache Lie algebras and introduce some basic
Smarandache properties about them. We do not promise to exhaust all notions about

Lie algebras and introduce them to Smarandache Lie algebras. Only those concepts
which has interested us and which we felt would be developable to Smarandache
concepts has been defined in this book.

D

EFINITION

4.2.1: A Smarandache real Lie algebra (S-real Lie algebra) L is a

vector space over the real field which has a proper subset X (X

⊂ L, X ≠ φ) such that

X is a Smarandache K- vectorial subspace of L furnished with a bilinear product [

µ] satisfying the identities [ λ , µ] = - [µ ,λ ], [[ λ , µ] , ν] + [[,µ, ν] , λ] +

[[

ν,λ],µ] = 0 for all λ , µ , ν ∈ X. It is important to note that this bilinear product

need not be defined on the totallity of L.

We always assume L to be a finite dimensional vector space over the field of reals.

This type of Smarandache Lie algebras will be called as Smarandache Lie algebras
of type I, but we will for short say Smarandache Lie algebras. We define type II and
other types in what follows. Similarly we define Smarandache complex Lie algebras
(S-complex Lie algebra). All real or complex Lie algebras need not in general be S-
real or complex Lie algebras.

T

HEOREM

4.2.1: Let L be a complex or a real Lie algebra; L is a S-complex or a

real Lie algebra if and only if L has a S-vectorial subspace X.

Proof: Straightforward from the very definitions and the fact that L is a complex or a
real Lie algebra so satisfies the identities.

Now we define the conditions under which a non-associative ring is a Lie algebra.

D

EFINITION

4.2.2: Let V be a non-associative algebra. V is said to be a

Smarandache Lie algebra (S-Lie algebra) if V has a proper subset W

⊂ V, W ≠ φ

such that W is a S-non-associative ring and every element of W satisfies the Lie
conditions x

= 0 and (xy)z + (yz) x + (zx) y =0 for all x, y, z

∈ W.

HEOREM

4.2.2: If V is a S-Lie algebra where V is a non-associative ring then V is a

SNA-ring.

Proof: Follows from the fact that if a non-associative ring has proper subset W,
where W is a SNA-ring then obviously by the very definition of SNA-ring, V is a
SNA-ring.

The following theorem is given as an exercise for the reader to prove.

T

HEOREM

4.2.3: If V is a SNA-ring then V in general need not be a S-Lie algebra

even if V is a Lie algebra.

Note: The reader must construct an example of a SNA-ring V which is a Lie ring
but V must not contain any SNA-subring.

Example 4.2.1: Let Z

= {0, 1, 2} be the prime field of characteristic three. G be a

groupoid given by the following table:

∗

G be the groupoid ring of G over Z

. W = {0, 2a

+ a

, 2a

}. Clearly W

satisfies the Lie identities and W is trivially a S-non- associative ring. Hence W is a
S-Lie ring but not a Lie ring.

Example 4.2.2: Let Z

= {0, 1, 2, 3} be the ring of characteristic 4 and G be the

groupoid given as in example 4.2.1. Clearly the groupoid ring is a S-Lie ring and not
a Lie ring. For take X = {0, 2a

+ 2a

, a

+ 3a

, 3a

+ a

Clearly Z

G is a S-Lie ring. But Z

G is never a Lie ring. In consequence of this we

have the following theorems, we in these examples have taken G

∈ Z

∗∗∗

(n).

HEOREM

4.2.44: Let F be any field or a commutative ring. FG the groupoid ring

where G

∈ Z∗∗∗(n). No groupoid ring in this class of groupoid rings is a Lie

algebra.

Proof: F is a field or a commutative ring. Let G

∈ Z

∗∗∗

(n) for some n. FG the

groupoid ring of G over F. If FG is to a Lie algebra then it should satisfy the
following two identities.

(1) x

= 0 for all x

∈ FG.

(2) (xy) z + (yz)x + (zx)y = 0.

Now let g = Z

(t, u); atleast one of t or u

≠ 0. Now x

= 0

⇒ tx + ux = 0 so that (t +

u)x = 0 for all x

∈ G is possible only if t + u = n.

Now (xy)z + (yz)x + (zx)y = (tx + uy)z + (ty + uz)x + (tz + ux)y = t

x + uty + uz +

y + utz + ux + t

z + tux + uy = 0.

(x + y + z) + ut (y + z + x) + u (x + y + z) = 0 so that (t

+ ut + u) (x + y + z) = 0.

Thus for this equation to be true we must have t

+ ut + u = 0, t (u +t) + u = 0 and u +

t = n are true if and only if u = 0 forcing t = 0. Hence no groupoid in the class of
groupoid Z

∗∗∗

(n) satisfies the Jacobi identity together with x

= 0. So no groupoid

ring using the class of groupoid from Z

∗∗∗

(n) is a Lie algebra.

HEOREM

4.2.5: Let R be a commutative ring with unit or a field, L

(m) be a loop

from the new class of loops L

The loop ring RL

(m) is never a Lie algebra.

Proof: Follows from the fact that if RL

(m) is to be Lie algebra then x

= 0 and

(xy)z + (yz)x + (zx)y = 0 for all x

∈ RL

(m). If x

∈ L

(m)

⊆ RL

(m) we know x

= 0 is impossible as in L

(m), x

= e that is every element is self inversed or of order

two where e is the identity in L

(m).

Now if (xy)z + (yz)x + (zx)y = 0 for some x, y, z L

(m), then we have

[my – (m – 1)x ]z + [mz – (m – 1) y]x + [mx – (m - 1)z]y

mz – (m – 1)[my – (m – 1) x] + mx – (m – 1)[mz – (m – 1)y]
+ my – (m – 1)[mx – (m – 1)z]

mz – m (m – 1) y + (m – 1)

x + mx – (m – 1) mz + (m – 1)

+ my – (m – 1) mx + (m – 1)

m (x + y + z) + (m – 1)

(x + y + z) – m (m – 1) (x + y + z)

[m + (m – 1)

– m (m – 1)] (x + y + z)

[ m + m

– 2m + 1 – m

+ m] [x + y + z]

1. (x + y + z ) = 0.

This may not be true for all x, y, z

∈ RL

(m). Thus no loop ring RL

(m) from the

new class of loops L

will lead to a Lie algebra. Thus from these theorems we see

that the loop rings and groupoid rings constructed by us from the new class of loops
or groupoids does not become Lie algebras. Thus they form a distinctly interesting
class of non associative rings. But now our study will be "Can ever these class of
non-associative rings have Smarandache Lie rings". For this we proceed to study
some examples.

These examples show that we can have some of the loop rings or groupoid rings
where the loop are from the class of new loops and groupoid rings which are from
the new class of groupoids can be Smarandache Lie rings when we take appropriate
fields Z

; p can be prime or a composite number.

HEOREM

4.2.6: Let Z

G be a groupoid ring of the groupoid G, G

∈ Z

∗∗∗

(n) and Z

be the field or ring of integers modulo m. Z

G can be a S-Lie ring but never a Lie

ring.

Proof: We have already shown that Z

G is never a Lie ring for all G

∈ Z

∗∗∗

(n). To

prove ZnG is a Lie ring it is sufficient if we illustrate by example.

Now we proceed on to show that in the groupoid ring Z

G where G

∈ Z

∗∗∗

(n) the

set U = {

α = Σα

Σα

= 0}; is a ring which is non-associative. Clearly no element

in U can be from G. Since m is a composite number or a prime different from 2 we
can have in some cases U to be a S-Lie ring, if elements of G satisfy the two
identities x

= 0 and (xy) z + (yz) x + (zx) y = 0 for x, y, z

∈ Z

Thus we have classes of non-associative rings which are never Lie rings but can be
S-Lie rings.

ROBLEMS

1. Is

G a S-Lie ring where G = Z

(3,0)?

2. Can

(3) be a S-Lie ring?

3. Let

(5) be a loop ring.

{

}

)

(

and

∈

∑

. Is

U a Smarandache NA ring? Does U satisfy Lie identity?

Find whether Z

G where G = Z

(5,4) be a S-Lie ring.

5. Show

(m) for n = 15, m = 14 is not a Lie ring.

Give an example of a S-Lie ring using Z

and a groupoid from Z

∗∗∗

(8).

7. Can

G for any G

∈ Z

∗∗∗

(n) be a S-Lie ring? Justify your answer.

8. Show

G for G

∈ Z

∗∗∗

(n) is a S-Lie ring.

9. Prove

(7) is never a Lie ring. Is it a S-Lie ring?

10. Show

(8) is not a S-Lie ring.

4.3 Substructures in Smarandache Lie algebras

In this section we introduce the concept of substructures like subalgebras, ideals and
quotient algebras in Lie algebras and also define their Smarandache analogue. Here
we introduce several new concepts in Lie algebras and Smarandache Lie algebras
which are not discussed or defined till date. Some of the concepts introduced in this
section are, Chinese Lie algebras and its Smarandache analogue, SI

∗

and I

∗ Lie

algebras, ideally obedient Lie algebras and Smarandache ideally obedient Lie
algebras CN-Lie algebras and Smarandache CN-Lie algebras. Finally the concepts
in Lie algebras like pseudo ideal, semiprime etc are introduced and their
Smarandache analogue also discussed.

D

EFINITION

4.3.1: Let L be S-Lie algebra. Suppose L has a subalgebra say X and X

is a S-Lie algebra then we call X the Smarandache Lie subalgebra of L (S-Lie
subalgebra).

T

HEOREM

4.3.1: Let L be a Lie algebra if L has a S-Lie subalgebra then L is itself a

S-Lie algebra.

Proof: Straightforward from the very definitions.

D

EFINITION

4.3.2: Let R and S be two non-associative rings, which are S-Lie

algebras. We say

φ: R → S is a Smarandache Lie algebra homomorphism (S-Lie

algebra homomorphism) if

φ restricted to the subsets W of R and V of S respectively

where W and V are SNA-subrings which satisfy the Lie identity and

φ: W → V is a

Lie algebra homomorphism.

Note: It is important to note that the mapping need not even be defined on the whole
of R it is sufficient if it is defined on W, W a proper subset which is a Lie algebra in
R. The concept of Smarandache Lie isomorphism (S-Lie isomorphism) is defined in
a similar way.

D

EFINITION

4.3.3: Let L be a Lie algebra. A subspace S of L is said to be a

subalgebra if [S, S]

⊂ S i.e. for x, y ∈ S, [x, y] ∈ S. A subspace S of L is said to be an

100

ideal if [L, S]

⊆ S i.e. x ∈ L and y ∈ S then [x, y] ∈ S. An ideal is a subalgebra. We

say X

⊂ L is a Smarandache Lie ideal (S-Lie ideal) of L if (1) X ⊂ Y is a

Smarandache Lie Algebra (S-Lie Algebra) where Y is a proper subspace of L which
satisfies the Lie identify for all x

∈ X and y ∈ Y, we have [x, y] ∈ X. If a Lie algebra

L has no Smarandache Lie ideals then we call L a Smarandache simple Lie algebra
(S-simple Lie algebra).

D

EFINITION

4.3.4: Let L be a real or complex vector space over a field K. A S-Lie

ideal X of L is said to be Smarandache Lie solvable (S-Lie solvable) if

⊃

X' = [X, X]

⊇

X'' = [X', X'] …

⊇

= [ X

k-1

, X

k-1

]

⊇ …

are Smarandache Lie ideals and X

= 0 for some positive integer k. Similarly we say

a S-Lie algebra L is Smarandache Lie solvable if, X the Smarandache K-vectorial
subspace of L is such that

⊇ X'

= [X, X]

⊇

X'' = [ X

, X

] …

⊇

= [ X

k-1

, X

k-1

]

⊇ …

are Smarandache K-vectorial subspaces and X

= 0 for some positive integer K.

EFINITION

4.3.5: Let L be a S-Lie algebra, X a S-Lie ideal of L; we say X is a

Smarandache minimal Lie ideal (S-minmal Lie ideal) of L if Y

⊂ X and Y is a S-Lie

ideal of X then X = Y or Y =

φ.

HEOREM

4.3.2: Let L be a S-Lie algebra X a S-minimal ideal of L then [X, X] = 0

or if X' = [X X]

⊂ X then X' is not a S-Lie ideal of L or any [X

, X

] is not a S-Lie

ideal.

Proof: Straightforward by the very definitions of these concepts.

D

EFINITION

4.3.6: Let L be a S-Lie algebra. Let B

⊂ L be a S-Lie ideal of L, we say

B is a Smarandache maximal Lie ideal (S-maximal Lie ideal) of L if L

⊃ A ⊃ B, A

any S-Lie ideal of L then B = A or A = L.

T

HEOREM

4.3.3: Let L be a Smarandache Lie algebra. If B is a S-maximal Lie ideal

of L and if [B B]

⊂ C then either C = L or C is not a S-Lie ideal of L.

Proof: As in the case of minimality, one can easily prove this result.

D

EFINITION

4.3.7: Let L be a S-Lie algebra we say the Smarandache D.C.C

(S.D.C.C) conditions on S-Lie ideals is satisfied by L if every descending chains X

⊃

⊃ … of S-Lie ideals of L is sationary. We say the S-Lie algebra satisfies

Smarandache A.C.C (S.A.C.C) condition on S-Lie ideals if every ascending chain Y

⊂ Y

⊂… of S-Lie ideals of L is stationary.

101

HEOREM

4.3.4: Let L be a S-finite dimensional Lie algebra. L satisfies S.A.C.C on

S- Lie ideals of L.

Proof: Straightforward by using the fact that

1. S-basis is finite.
2. All ideals X are basically subspaces of L i.e. [X X]

⊂ L.

EFINITION

4.3.8: Let L be a vector space over the field K. L be a S-Lie algebra. X

be a subspace of L i.e. X is only a subspace and not a Lie subspace of L. We call X a
Smarandache pseudo Lie ideal (S-pseudo Lie ideal) of L if for every y

∈ Y where Y is

a subset of L and a K-algebra of L, which satisfies the Lie identity and we have [xy]
∈ X for all x ∈ X.

Note: [X, X]

⊄ X i.e. [x, x'] ∉ [X, X] for some x, x' ∈ X.

Clearly S-pseudo Lie ideals are in general not S-Lie ideals of L. The study of
characterization of Smarandache pseudo Lie ideals is left for the reader as an open
problem. Any researcher can define Smarandache pseudo minimal Lie ideals and
Smarandache pseudo maximal Lie ideals using the usual notions of minimality and
maximality. We say a S-Lie ring L is S-pseudo Lie simple if L has no proper
Smarandache pseudo Lie ideals (S-pseudo Lie ideals).

D

EFINITION

4.3.9: Let L be a S-Lie ring, X a S-pseudo Lie ideal of L then L/X = {a +

 a ∈ L} is defined as the Smarandache pseudo quotient ring (S-pseudo quotient

ring).

The structure possessed by L / X is an interesting problem for it depends both on the
properties enjoyed by L and that of X.

D

EFINITION

4.3.10: Let L be a Lie algebra. X be a proper subset which is a K-

algebra and satisfies the Lie identity. We say the Lie algebra L is Smarandache semi
prime (S-semi prime) if and only if L has no non-zero S-Lie ideal A with [A A] = 0.

T

HEOREM

4.3.5: Let L be a Lie ring. If L is S-semi prime then L has no S-minimal

ideals.

Proof: Straightforward from the definitions and earlier results.

D

EFINITION

4.3.11: Let L be a Lie ring. X a proper S-Lie subring of L; i.e. X is a

proper subset which is S-Lie subring. A Smarandache subsemi Lie ideal (S-subsemi
Lie ideal) I related to X i.e. I is a S-Lie ideal related to X. and I is not a S-Lie ideal
related to any other S-subring containing X.

If L has a S-subsemiideal then we say L is a Smarandache subsemi Lie ideal ring (S-
subsemi Lie ideal ring). We call the Lie ring a Smarandache filial Lie ring (S-filial
Lie ring) if the relation S-Lie ideal in L is transitive that is a S-subring J is a S-Lie
ideal in S-subring I and I a S-Lie ideal in L then J is an S-ideal of L.

Such study has not been imitated till date.

102

EFINITION

4.3.12: Let L be a Lie algebra. We say the Lie algebra is a Chinese Lie

algebra, if given elements a, b

∈ L and ideals I and J in L such that [J + I, a] = [I

+ J, b], there exists c

∈ L such that [I, a] = [I, c] and [J, c] = [J, b]; L will be called

Smarandache Chinese Lie ring (S-Chinese Lie ring) if we replace the ideals I and J
of L by S-Lie ideals.

D

EFINITION

9.3.13: Let L be a Lie algebra. An ideal I of L is said to be a Lie

multiplication ideal J

⊂ I if J = [I, C] for some ideal C. We call the multiplication as

Smarandache Lie multiplication ideal (S-Lie multiplication ideal) if for S-Lie ideal I
of L we have J

⊂ I (J also a S- Lie ideal) such that J = [I,C] where C is a S-Lie ideal.

EFINITION

4.3.14: Let L be a Lie algebra. We call I a Lie obedient ideal of L if we

have two ideals X, Y in L, X

≠ Y such that [X ∩ I, Y ∩ I] = [X, Y] ∩ I. We say I is a

Smarandache Lie obedient ideal (S-Lie obedient ideal) of L, if I is a S-Lie ideal and
X and Y X

≠ Y are also S-Lie ideals satisfying the relation [X ∩ I, Y ∩ I] = [X, Y]

∩ I. If every ideal is Lie obedient we call the Lie ring an ideally obedient Lie ring.
If every S-Lie ideal is Smarandache Lie obedient we call L a Smarandache ideally
obedient Lie ring (S-ideally obedient Lie ring).

D

EFINITION

4.3.15: Let L be a Lie algebra. L is said to be a I

∗ - ring if for every

pair of ideals I

, I

∈ {I

}, where I

denotes the collection of all ideals of L

and for

every a

∈ L \ (I

∪ I

) we have [a, I

] = [a, I

] is an ideal of {I

]. Suppose for the Lie

algebra L if {

s
j

I } denotes the collection of S-Lie ideals of L we call L a SI∗ ring if for

every a

∈ L \ (I

∪ I

) we have [a, I

] = [a, I

] is an ideal of {

s
j

I }.

EFINITION

4.3.16: Let L be a Lie algebra we say L is locally unitary if for each x

∈

L there exists an idempotent e

∈ L such that xe = ex = x . We call the Lie algebra L

Smarandache locally semi unitary (S-locally semi unitary) if for every x

∈ L there

exists a S-idempotent e in L such that x e = e x = x.

D

EFINITION

4.3.17: Let L be a Lie algebra. A subset P of L is called a closed net if

1) P is a semi group under ‘+’ and ‘.’.
2) aP

⊂ P for all a ∈ K .

(K is the field over which L is defined). We call P a Smarandache closed net (S-
closed net) if P is a S-semi group under ‘.’ and a P

⊂ P for all a∈ K.

EFINITION

4.3.18:

We call the Lie algebra L a CN-ring (closed net ring) if L =

∪ P

where P

‘ s are closed nets in L such that P

∩ P

φ or {1} or {0} if i ≠ j and P

∩

= P

if i = j (i.e. each P

is nontrivial).

We call the Lie algebra a weakly CN ring or algebra if L

⊂ ∪ P

with P

∩ P

≠ φ or

{1} or {0} even if i

≠ j.

EFINITION

4.3.19: Let L be a Lie algebra {SP

} denote the collection of all S-closed

nets of L. We say L is a Smarandache CN-ring (S-CN-ring) if L =

∪ SP

and SP

∩

103

φ or {0} or {1} if i ≠ j and SP

∩ SP

= SP

if i = j. We say L is a Smarandache

weakly CN-ring (S-weakly CN-ring) if L

⊂ ∪ {SP

} and SP

∩ SP

≠ φ or {0} or {1}.

HEOREM

4.3.6: Let L be a S-CN ring then L is a S-weakly CN-ring.

Proof: Follows from the very definition.

The proof of the following theorem is left for the reader.

T

HEOREM

4.3.7:

Every S-weakly CN ring in general is not a S-CN-ring.

ROBLEMS

Give an example of a Lie algebra, which is not a S-Lie algebra.

Find S-ideals I in any Lie algebra and show I is not an ideal.

Find a Lie ring, which has no S-pseudo ideals.

Give an example of a Lie algebra, which has S-pseudo ideals.

Give an example of a S-Chinese Lie ring.

Prove all algebras are not S-Chinese Lie algebras by an example.

Give an example of a S-weakly CN-ring which is not a S-CN-ring

Give an example of a Lie algebra, which has no S-obedient Lie ideals.

Give an example of a S-ideally obedient ring.

10.

Find examples of semi prime Lie rings and S-semi prime Lie rings.

11.

Find S-ideals in a Lie ring L, which are not ideals of L.

12.

Illustrate by examples: Lie multiplication ideal and S-Lie multiplication
ideal of a Lie algebra L.

4.4 Special Properties in Smarandache Lie Algebras

In this section we introduce several properties about elements in Lie algebras and
their Smarandache analogous. All the properties introduced in this section are very
new and interesting. A deep study of these definitions and notions will certainly lead
to more innovative and unraveled properties about Lie algebras and in particular of
S-Lie algebras. The main new concepts defined in this section are Smarandache Lie
related element, Smarandache Lie links, essential Lie algebra and its Smarandache
analogue. Stabilizer and stable Lie algebra and its Smarandache notion Lie super ore
conditions, conjugate Lie subalgebras and Smarandache conjugate Lie subalgebras
and finally the concept of Smarandache generalized stable Lie algebras.

These notions are defined using mainly the Lie product. To the best of our
knowledge so far no one has explored or defined these new concepts in Lie algebras,
hence cannot be found in S-Lie algebras.

D

EFINITION

4.4.1: Let L be a Lie algebra. We say any two elements a, b

∈ L are Lie

related if there exists a subspace A of L such that a, b

∉ A but a ∈ [A, b] and b ∈ [A,

a]. We say a, b

∈ L are Smarandache Lie related (S-Lie related) if there exists a S-

subspace A of L such that a, b

∉ A but a ∈ [A, b] and b ∈ [A, a]. We say a and b are

Smarandache Lie link related elements (S-Lie link related elements) by the link A.

104

EFINITION

4.4.2: Let L be a Lie algebra. A subalgebra B of L is said to be an

essential algebra of L if the intersection of B with every other subalgebra is zero. (By
subalgebras we mean only proper subalgebras).

A S-subalgebra A of L is said to be a Smarandache essential algebra (S-essential
algebra) if the intersection of A with every other S-subalgebra of L is zero. If every
subalgebra of a Lie algebra is essential then we call L a essential Lie ring. If every
S-subalgebra of a Lie algebra is S-essential then L is a Smarandache essential Lie
algebra (S-essential Lie algebra).

D

EFINITION

4.4.3: Let L be a Lie algebra. If for a pair of subalgebras P and Q of L

there exists a subalgebra T of L (T

≠ L ) such that [P, T] and [T, Q] are subrings

with [P, T] = [T, Q]; then we say the pair of subalgebras are stabilized subalgebras
and T is called the stabilizer subalgebra of P and Q.

We say a pair of S- subalgebras P and Q of L are Smarandache stabilized algebras
(S-stabilized algebras) if there exists a S-subalgebra T such that [P, T] = [T, Q] are
S-subalgebras and T is called the Smarandache stabilized subalgebra (S-stabilized
subalgebra) of P and Q.

It is left for the reader to prove the following theporem by constructing an example:

T

HEOREM

4.4.1: Let L be a Lie ring. If the subalgebras A, B of L is a stable pair

then A, B is a stabilized pair and not conversely. The same result holds good for
Smarandache stable pair of S-subalgebras.

D

EFINITION

4.4.4: Let L be a Lie algebras if every pair of subalgebras of L is a

stable pair, then we say the Lie algebra is a stable algebra. The Lie algebra is a
Smarandache stable algebra (S-stable algebra) if every pair of S-subalgebras of L is
a Smarandache stable pair (S-stable pair) then we say the Lie algebra is a
Smarandache stable algebra (S-stable algebra).

T

HEOREM

4.4.2: Every S-stable Lie algebra is a S-stabilized Lie algebra.

Proof: Follows from the very definition hence left for the reader to prove.

D

EFINITION

4.4.5: Let L be a Lie algebra. We say a pair of elements r, s

∈ L

satisfies the Lie super ore condition if there is some r'

∈ L such that [r, s] = [s, r']. If

every pair of elements satisfies the Lie super one condition then we say L satisfies the
Lie super ore condition.

D

EFINITION

4.4.6: Let L be a Lie algebra such that L is a S-Lie algebra i.e. X

⊂ L is

such that X is a S-K-vectorial subspace satisfying the Lie identity. We say a pair of
elements x , y

∈ X satisfies Smarandache super ore condition (S-super ore condition)

if there is some z

∈ L such That [x, y] = [y, z ]. If every pair of elements in X satisfies

S-super ore condition then we say L satisfies the Smarandache super ore condition
(S-super ore condition).

T

HEOREM

4.4.3: Let L be a Lie algebra if L satisfies the S-super ore condition then

L need not in general be a Lie algebra satisfying super ore condition.

105

Proof: The reader is requested to construct an example to prove the theorem which is
direct.

T

HEOREM

4.4.4: If L is a Lie algebra which satisfies the super ore condition and if L

is a S-Lie algebra then L satisfies S-super ore condition.

Proof: Straightforward by the very definition.

D

EFINITION

4.4.7: Let L be a Lie ring. Two subalgebras A and B of same dimension

are conjugate if there exists some x

∈ L such that Ax = xB (xA = By).

We call L a conjugate Lie algebra if every distinct pair of subalgebras of same
dimension are conjugate.

Now we define it to the case of Smarandache conjugates.

D

EFINITION

4.4.8: Let L be a Lie algebra. Two S-subalgebras A and B of same

Smarandache dimension are said to be Smarandache conjugate (S-conjugate) if
there exists x

∈ L such that Ax = xB (or xA = Bx). If every pair of S-subalgebras of

same Smarandache dimension are S-conjugate then we call L a Smarandache
conjugate Lie ring (algebra) (S-co-Lie-ring).

We call L a Smarandache weak Lie conjugate ring (S-weak co-Lie ring) if there
exists at least one pair of distinct S-subalgebras with same Smarandache dimension
which are conjugate to each other.

T

HEOREM

4.4.5: Every S-co-Lie-ring is a S-weak co-ring.

Proof: Direct by the very definitions.

D

EFINITION

4.4.9: Let L be a Lie algebra and P be a additive subgroup of L. P is

called the n capacitor group of L if [xP]

⊂ P for every x ∈ L. If in case of P an S-

semigroup and L a S-Lie algebra having X

⊂ L where X is a K-algebra of L then we

call P a Smarandache n-capacitor group (S-n-capacitor group) of L if [x, P]

⊂ P for

every x

∈ X.

HEOREM

4.4.6: Let L be a Lie algebra. Suppose I is an ideal of L then I is a n-

capacitor group of L.

Proof: Straightforward.

T

HEOREM

4.4.7: Let L be a Lie algebra if I is a S-ideal of L then I is a S-n-capacitor

group of L.

Proof: Direct by the very definitions.

D

EFINITION

4.4.10: Let L be a Lie algebra. A subset I of L which is closed under ‘+’

is called a n-ring ideal of L if [r, i]

⊂ I for all i ∈ I and r ∈ L. We say the subset I of

L which is closed under ‘+’ and has a subgroup under ‘+’ i.e. I is a S-semigroup to

106

be a Smarandache ring n-ideal (S-ring n-ideal) of L if L is a S-Lie algebra and r

∈ X

⊂ L where X is a proper subset which is a K-algebra.

D

EFINITION

4.4.11: Let L be a Lie ring or a Lie algebra. We define P (L) = {x

∈ L /

there exist s, t in L such that s(xt) = 1 or (sx)t = 1 }. L is called the generalized stable
Lie algebra provided that [a, P(L) ] + [b, P(L)] = P(L) with a, b

∈ L implies a + by

∈ P(L) for some y ∈ L.

If L is a S-Lie algebra with X a non-empty subset of L with X a K-algebra and SP(L)
= {x

∈ X / there exists s, t in L such that s(xt) = 1 or (sx) t = 1}. L is called the

Smarandache generalized stable Lie algebra (S-generalized stable Lie algebra)
provided [a, SP(L)] + [b , SP(L)] = SP(L) with a, b

∈ X ; a + by ∈ SP(L) for some y

∈ X.

D

EFINITION

4.4.12: Let L be a Lie algebra. L is said to satisfy the n-stable range

conditions for whenever a

, a

,…a

n+1

∈ L with [a

, L] + [a

, L] + …+ [a

, L] = L

there exists b

, b

, …, b

∈ L

such that [a

+ a

n+1

, L] + [a

+ a

n+1

, L] + …+

+ a

n+1

, b

, L] = L.

If L is a S-Lie algebra and X

⊂ L is a proper subset which is a K-vectorial algebra

and satisfies the Lie identity and b

, b

, …, b

∈ X and a

, …, a

n+1

∈ L with [a

, X ]

+…+ [a

n+1

, X] = X and [a

+ a

n+1

, X] +…+ [ a

+ a

n+1

, X] = X. Then L is

said to satisfy Smarandache n-stable range conditions (S-n-stable range conditions).

P

ROBLEMS

1.  Give an example of a Lie algebra which satisfies the Lie super ore condition.
2.  Give an example of a stable Lie algebra and S-stable S-Lie algebra.
3.  Construct a Lie algebra which are conjugate subalgebras but has no S-

conjugate subalgebras.

4. Give an example of a Lie algebra which is a n-ring ideal and a S-n-ring ideal.
5. Does there exist a generalized stable Lie algebra? a S-generalized stable Lie

algebra?

4.5 Some New Notions on Smarandache Lie Algebras

In this section we define Smarandache mixed direct product of Lie algebras. Only
this concept paves way to define different types of S-Lie algebras and study them.
Here notions of Smarandache simple, Smarandache semi-simple, Smarandache
strongly simple, Smarandache strongly semi-simple are introduced. The two types of
S-mixed direct product of Lie algebras helps us to see simultaneously certain special
structures like S-simple and S-semisimple to be present in them.

D

EFINITION

4.5.1: Let SL

= L

× … × L

are Lie algebras defined over the same

field K where at least one of the Lie algebras L

is the linear algebra over K. Then

is a direct product of Lie algebras where the Lie conditions are defined

component wise for the n-tuples. As they are defined on the same field
compactability exist. For example [(x

, …, x

), (y

, …, y

)] = ([x

], [x

], …,

107

]) where x

, y

∈ L

for i = 1, 2, …, n. SL

is called the Smarandache mixed

direct product (S-mixed direct product) of Lie algebras of type A.

T

HEOREM

4.5.1: Let SL

be the S-mixed direct product Lie algebra of type A. Then

is a S-Lie algebra.

Proof: Follows from the fact that SL

contains a Lie algebra L

which is K-algebra.

EFINITION

4.5.2: Let L be a S-Lie algebra if L has no non-zero S-solvable Lie ideal

then we say L is Smarandache semi-simple (S-semi simple). If L has no S-Lie ideals
other than 0 and L we call L, Smarandache simple (S-simple).

D

EFINITION

4.5.3: Let L be a Lie algebra we call L a Smarandache Lie simple (S-Lie

simple) if L has a S-subalgebra which is S-simple (i.e. if X is a S-subalgebra of L
and X has no S-ideals other than 0 and X).

D

EFINITION

4.5.4: Let L be a Lie algebra, we call L a Smarandache Lie semi-simple

(S-Lie semi-simple) if L has a S-subalgebra X such that X is S-semi-simple i.e. X is a
S-subalgebra of L and X has no non-zero solvable S-Lie ideal.

D

EFINITION

4.5.5: Let L be a Lie algebra. If every proper S-subalgebra X of L is S-

Lie simple we say the Lie algebra is Smarandache strongly Lie simple (S-strongly
Lie simple).

T

HEOREM

4.5.2: If L is S-strongly Lie simple then L is S-Lie simple.

Proof: Direct by the very definitions of these concepts.

D

EFINITION

4.5.6: Let L be a Lie algebra if every S-subalgebra X of L is S-Lie semi-

simple then we call L Smarandache strongly Lie semi-simple (S-strongly Lie semi-
simple).

Consequent of these definitions we have the following theorem which is left as an
exercise for the reader to prove.

T

HEOREM

4.5.3: Let L be a Lie algebra which is S-strongly Lie semi-simple then L is

S-Lie semi-simple.

Now we proceed on to define S-mixed direct product of type B.

D

EFINITION

4.5.7: Let SL

= L

× … × L

be the direct product of Lie algebras L

…, L

defined over the same field K, in which at least one of the L

is a linear algebra

over K and is a S-semi simple Lie Algebra and one of the L

's ( i

≠ j) is a linear

algebra over K and is S-simple Lie algebra, then we call SL

the Smarandache

mixed direct product of Lie algebras of type B.

T

HEOREM

4.5.4: Let SL

be the S-mixed direct product of Lie algebras of type B then

a. S-Lie semi-simple.
b. S-Lie simple.

108

Proof: Follows directly by the very definitions.

Thus we see that in case of Smarandache structures only we can have simultaneously
the Lie algebra to be S-Lie simple and S-Lie semi-simple. Thus this definition of S-
mixed direct product of type B helps us to find algebras which is both S-Lie simple
and S-Lie semi-simple.

D

EFINITION

4.5.8: Let L be a Lie algebra. We call L a Smarandache sympletic Lie

algebra (S-sympletic Lie algebra) if L has a S-subalgebra X, which is a sympletic Lie
algebra.

Similarly one can define Smarandache triangular (S-triangular) and Smarandache
orthogonal (S-orthogonal) Lie algebras.

D

EFINITION

4.5.9: Let L be a Lie algebra. X be a S-subalgebra of L. The

Smarandache normalizer (S-normalizer) SN(X) = {x

∈ L / [x, X] ⊆ X} that is [x, y] ∈

X for every y

∈ X. SN(X) is a subalgebra. We call SN(X) the Smarandache

normalizer only when SN(X) is a S-subalgebra.

D

EFINITION

4.5.10: A S-subalgebra X of a Lie algebra L is called a Smarandache

Cartan subalgebra (S-Cartan subalgebra) if

a. X is nilpotent.
b. X is its own S-normalizer in L.

Thus the Smarandache mixed direct products of Lie algebras of type A and B leads
us to construct more and more examples of S-Lie algebras, S-Lie simple algebras
and S-Lie semi-simple algebras. Further S-strong simple and semi-simple algebras
have been also introduced. So given any Lie algebra which is not a S-Lie algebra can
be made into a S-Lie algebra by S-mixed direct product of Lie algebras.

P

ROBLEMS

1.  Give an example of a S- Lie simple algebra.
2.  Illustrate by an example the notion of S-strong Lie semi-simple algebra.
3.  Give an example of a Lie algebra which is both S-Lie simple and S-Lie semi-

simple.

4. Show every Cartan subalgebra need not be a S-Cartan subalgebra.
5. Give an example of S-mixed direct product which is a S-strong semi-simple

algebra.

6. Can type B S-mixed direct product of Lie algebras be S-strong simple Lie

algebras? Justify your claim.

7. Using S-mixed direct product of Lie algebras construct a S-Cartan Lie

algebra.

109

Chapter 5

JORDAN ALGEBRAS AND
SMARANDACHE JORDAN ALGEBRAS

This chapter has two sections.  The first section is devoted to the recollection and
introduction of basic concepts about Jordan algebras. As the class of Jordan algebras
is a large one and there are good books which give a complete and an excellent
exposition about Jordan algebras [2, 37, 51] we don't intend going deep into the
study. As the main aim of this book is only to introduce Smarandache non-
associative algebras we don't give a complete analysis of Jordan algebras. Further we
are mainly interested in the algebraic aspect of it and not view it as a vector space or
its associated transformations. We describe and study chiefly Jordan algebras as non-
associative algebras which satisfies the Jordan identity.

Section 2 of this chapter introduces the concept of Smarandache Jordan algebras. We
introduce several properties enjoyed by associative rings some of them have been
modified and incorporated as Smarandache notions in S-Jordan algebras. To the best
of our knowledge such study of introduction of these concepts in Jordan algebras is
totally absent. Further all books on non-associative algebras which deals with Jordan
algebras studies Jordan algebras in a very different angle. Here our study of
Smarandache Jordan algebra in this book is varied and makes an interesting one as
the notions/ concepts enjoyed by associative algebras cannot be totally deprived from
our study when we assume all the algebraic structure remains the same, except the
associativity.

So keeping this in mind we have introduced several new notions like Smarandache
Jordan strong ideal property, Smarandache power-joined algebras, Smarandache
magnifying and Smarandache shrinking elements in Jordan algebras and other
Smarandache concepts.

We have left for the reader to develop each of these concepts and obtain many
innovative results about Smarandache strong Jordan algebras, Smarandache
commutative Jordan algebra, Smarandache weakly commutative Jordan algebras,
Smarandache pseudo commutative Jordan algebras, Smarandache quasi commutative
Jordan algebra, Smarandache reduced Jordan algebras, Smarandache periodic Jordan
algebras, Smarandache E-Jordan algebras, and Smarandache pre-Boolean algebras.
The sole aim of this text is to give innovative and new concepts and notions which
can be developed into a very good theory.

5.1 Basic properties of Jordan Algebras

In this section we just recall the definition of Jordan algebras and derive some of its
basic properties. Several properties about Jordan algebras and its substructures like
radical and ideal are given. Examples are also given. For more about Jordan algebras
please refer [2, 37, 51].

110

EFINITION

[51]: Let A be a finite dimensional commutative algebra with identity. A

is not assumed to be associative. A is called a Jordan algebra if we have x

(xy) =

x(x

y) for all x, y

∈ A. An equivalent definition is A is a Jordan algebra if and only if

we have for all invertible x

∈ A and for all y ∈ A, x

–1

(xy) = x(x

–1

y).

EFINITION

5.1.1: A quadratic Jordan algebra is a triple J = (V, P, e) where V is a

finite dimensional vector space over K, P a quadratic map of V into the space end(V)
of its endomorphisms and e a non-zero element of V such that the axioms to be stated
below hold. Putting P(x, y) = P(x + y) – P(x) – P(y) we obtain a symmetric bilinear
map P, of V

× V into End V. The axioms for a quadratic Jordan algebras are as

follows:

1.  P(e) = e, P(x, e)y = P(x, y)e.
2.  P(P(x)y) = P(x)P(y)P(x).
3.  P(x)P(y, z)x = P(P(x)y, x)z.

J is said to be defined over k

⊂ K if there exists a k-structure on the vector space V

such that P is defined over k and that e

∈ V(k). J = (V, P, e) is a quadratic Jordan

algebra.

For more properties and results please refer [51] . We do not approach or define
Jordan algebras using Zaraski topology. We purely deal it as an algebraic structure or
to be more precise as an algebra i.e. like an algebra satisfying the identity a

(au) =

a(a

u).

EFINITION

5.1.2: Let A and B be any two Jordan algebras where we assume they

are finite dimensional over a field K of characteristic not equal to two. A linear
mapping

σ : A → B is called a Jordan homomorphism if

σ(a

) = (

σ(a))

for every a

∈ A.

ii.

σ(a.b.a) = σ(a)σ(b)σ(a) for every a, b ∈ A.

If B does not contain a zero divisor then

σ(a . b) = σ(a) . σ(b) or σ(a . b) = σ(b) .

σ(a).

We can also derive or define special Jordan algebras using an associative algebra
A. Let A be an associative algebra, Define a new multiplication '.' in A by a . b = (ab
+ ba)/2. We then have a Jordan algebra A

. A subalgebra of the Jordan algebra A

is called the special Jordan algebra. We just recall the notion of free special Jordan
algebra. Let K[x

, …, x

] be the non-commutative free-ring in the indeterminates x

…, x

(that is K[x

, …, x

] is the associative algebra over K that has as its K-basis

the free semigroup with identity element 1 over the free generators x

, x

, …, x

. The

subalgebra K[x

1, …,

]

generated by 1 and the x

's is called the free special Jordan

algebra of n-generators and is denoted by J

(n)

. A Jordan algebra A is special if and

only if there is an isomorphism from A onto B

, where B is some associative

algebra. A Jordan algebra that is not special is called exceptional.

A Jordan algebra A has a unique solvable ideal N, which contains all nilpotent
ideals of A and is called the radical of A. If N = 0, A is called semisimple. The

111

quotient A/N is always semisimple. A semisimple Jordan algebra A contains the unit
element and can be decomposed into a direct sum, A = A

⊕ … ⊕ A

of minimal

ideals A

, each A

is a simple algebra. In particular if K is of characteristic zero there

is a semisimple subalgebra S of A such that A = S

⊕ N. Let e be an idempotent

element of A and let

λ ∈ K. Put A

(

λ) = {x / x ∈ A, e . x = λx}. Then we have A =

(1)

⊕ A

(1/2)

⊕ A

(0). This decomposition is called the Pierce decomposition of A

relative to e. Suppose that the unity element 1 is expressed as a sum of the mutually
orthogonal idempotents of e.

A representation S of a Jordan algebra A on a K-linear space M is a K-linear
mapping a

→ S

from A into the associative algebra E(M) of all K-endomorphisms

of M such that

1. [S

, S

b.c

] + [S

, S

c.a

] + [S

, S

a.b

] = 0 and

2. S

+ S

= S

b.c

+ S

a.b

for all a, b, c in A.

A K-linear space M is called a Jordan module of A if there are given bilinear
mappings M

× A → M (denoted by (x, a) → x.a), A × M → M (denoted by (a, x) →

ax) such that for every x

∈ M and for every a, b, c ∈ A

x . a = a . x.

ii.

(x . a) . (b . c) + (x . b) (a . c) + (x . c )(a . b) = (x . (b . c)) . a .
+ (x . (a. c)) . b + (x . (a . b)) . c and

iii. x . a . b . c + x . c . b . a + x . a . c . b = (x . c)(a . b) + (x . a)(b . c)

+ (x . a)(a . c).

As usual there is a natural bijection between the representations of A and the Jordan
modules of A. A special representation of a Jordan algebra A is a homomorphism A
→ E

, where E is an associative algebra. Among the special representations of A

there exists a unique universal one in the following sense. There exists a special
representation S : A

→ U

with the following property. For every special

representation

σ : A → E

there exists a unique homomorphism

η : U

→ E

such

that

σ = ηS. The pair (U, S) is uniquely determined. Further more, if A is n-

dimensional over K, U is of dimension













over K. The pair (U, S) is called the

special universal enveloping algebra of A. A is special if and only if S : A

→ U

injective.

For more about Jordan algebras please refer [2, 37, 51].

Now we proceed on to study which of the loop rings using the new class of loops are
Jordan algebras and which of the groupoid rings are Jordan algebras. We first recall
the definition of Jordan loop [66]

D

EFINITION

[66]: Let (L, ., e) be a loop. We say L is a Jordan loop if

ab = ba.

ii)

(ab) = a(a

b) for all a, b

∈ L.

112

Example 5.1.1: Let L = {e, g

, g

, …, g

} be a Jordan loop given by the following

table:

. e g

e e g

e g

Clearly g

= g

and a

(ab) = a(a

b) for all a, b

∈ L = {e, g

, …, g

}. We have a

class of loops of even order which forms a Jordan loop.

D

EFINITION

[66]: Let J

= {e, g

, …, g

} where p is a prime greater than 3. Define '

∗'

on J

a. g

∗ g

= e.

b. e

∗ g

= g

∗ e = g

(e identity element of J

c. g

∗ g

= g

where

t =

)

(mod

)

(

)

(









−

for all g

, g

∈ J

, g

≠ g

. Clearly J

is a loop which is easily verified to be a Jordan

loop and order of J

is even. C(J

) = {J

/ p is an odd prime greater than 3} forms a

class of Jordan loops of even order.

T

HEOREM

[66]: Let Z

= {0, 1} be the prime field of characteristic two and J

be a

loop. The Jordan loop ring Z

is a Jordan ring.

Proof: Given J

= {e, g

, …, g

}, p > 3, p a prime. Z

= {0, 1} be the field of

characteristic two. Z

is a loop-ring. Every element

α ∈ Z

is of the form

= 0

or 1 according as m is even or odd (where

α =

∑

g i.e. m is the number of terms

which are non-zero in

α). Suppose α

= 0, then for any

β we have α

(

αβ) = 0 and

α(α

β) = 0. Thus α

(

αβ) = α(α

β). Suppose α

= 1, then for any

β ∈ Z

L we have

(

αβ) = αβ and α(α

β) = αβ. Thus once again α

(

αβ) = α(α

β). Hence Z

is a

Jordan loop ring. Thus we have got a class of Jordan rings of varying order given by
Z

where J

∈ C(J

HEOREM

[66]: Let K be a field of characterisitic zero and J

∈ C(J

). The loop ring

is a Jordan ring.

113

Proof: Follows from the fact that every

α, β ∈ KJ

we have

(

αβ) = α(α

β) as

every pair of elements in J

satisfies the identity a

(ab) = a(a

b) for every a, b

∈ J

Now we recall a groupoid in the class of groupoids Z

∗∗∗

(n) to be a Jordan groupoid.

EFINITION

5.1.3: A groupoid (G,

∗) is said to be Jordan groupoid if

a. ab = ba.
b. a

(ab) = a(a

for all a, b

∈ G.

HEOREM

5.1.1: Let G be a commutative groupoid with 1 in which every g

∈ G is

such that g

= 1. F be any field. Then the groupoid ring FG is a Jordan ring.

Proof: Straightforward by the definitions and simple calculations.

Thus we see groupoid rings and loop rings can also be Jordan rings for suitable loops
and groupoids.

D

EFINITION

[37]: An element a of a Jordan algebra is called quasi invertible if there

exists b in the algebra such that a + b – a . b = 0 and a + b – 2a . b – a

+ a

. b = 0.

Then b is called the quasi inverse of a. For algebras with 1 this is equivalent to 1 – a
is invertible with inverse 1 – b.

Further no idempotent is quasi invertible and every nilpotent element is quasi
invertible. Further it has been proved that for an algebraic Jordan algebra the
following two conditions are equivalent:

a. Every element is nilpotent.
b. Every element is quasi invertible.

5.2 Smarandache Jordan Algebras and its basic properties

Throughout this section we assume the Jordan algebra to be an algebraic Jordan
algebra A, i.e, a Jordan algebra is one which satisfies the identity x(x

y) = x

(xy), x,

∈ A. If A is commutative then ab = ba, otherwise we say A is a non-commutative

Jordan algebra. A is assumed to contain the multiplicative identity 1, i.e. A is
assumed to be a finite dimensional algebra with 1. We are only interested in
analyzing its algebraic aspects and we do not see its properties via any other
mathematical structure. All the more we do not even deal it as a matrix structure. For
the main reason is that Smarandache matrix theory and Smarandache vector spaces
and Smarandache linear algebras are yet to be completely developed into solid
theories.

D

EFINITION

5.2.1: Let A be any algebra. We say A is a Smarandache Jordan algebra

(S-Jordan algebra) if A has a proper subset X which is a subalgebra with respect to
the operation of A and X satisfies the identity x(x

y) = x

(xy) for all x, y

∈ X.

114

HEOREM

5.2.1: Let A be a Jordan algebra then A is a S-Jordan algebra.

Proof: Direct by the very definition.

T

HEOREM

5.2.2: Let A be a S-Jordan algebra then A need not be a Jordan algebra.

Proof: By an example.

Let Z

L be the loop algebra of a loop L over Z

. Here L is a loop having a subloop H

such that H is commutative and every element in H is such that g

= 1 for every g

∈

H. Clearly the loop algebra Z

G has Z

H as a subalgebra and Z

H satisfies the Jordan

identity. x(x

y) = x

(xy) for all x, y

∈ Z

Thus Z

is not a Jordan algebra but Z

H is a Jordan algebra, hence Z

G is a S-

Jordan algebra.

D

EFINITION

5.2.2: Let A be a algebra. If every subalgebra of A satisfies the Jordan

identity then we call A the Smarandache strong Jordan algebra (S-strong Jordan
algebra).

T

HEOREM

5.2.3: If A is a S-strong Jordan algebra then A is a S-Jordan algebra.

Proof: Follows by the very definitions directly.

The reader is expected to prove the following theorem:

T

HEOREM

5.2.4: If A is a S-Jordan algebra then A need not in general be a S-strong

Jordan algebra.

We derive and define several algebraic properties and their Smarandache analogue in
case of Jordan algebras. We see in case of Jordan algebras the concept of S-
idempotents are defined analogous to S-idempotents in usual associative algebras.
We define several nice and new notions and their Smarandache analogue.

D

EFINITION

5.2.3: Let A be an algebra. We say a proper subset B of A to be a

Smarandache subJordan algebra (S-subJordan algebra) if B

⊂ X, X ⊂ A, X is a S-

Jordan algebra.

D

EFINITION

5.2.4: Let A be an algebra which is S-Jordan algebra. We say A is a

Smarandache commutative Jordan algebra (S-commutative Jordan algebra) if every
subalgebra is commutative, We call A to be a Smarandache weakly commutative
Jordan algebra (S-weakly commutative Jordan algebra) if at least one of the
subalgebra which is a S-Jordan algebra is commutative.

T

HEOREM

5.2.5: Every S commutative Jordan algebra is S-weakly commutative

Jordan algebra.

Proof: Follows directly by the very definitions.

115

HEOREM

5.2.6: Let A be an S-commutative Jordan algebra. A need not be a

commutative algebra.

Proof: The reader is requested to construct an example to prove the theorem.

We define Smarandache ideals in algebras only algebraically.

D

EFINITION

5.2.5: Let A be an algebra. X

⊂ A be a subset which is a S-Jordan

algebra i.e. X satisfies the Jordan identity. A proper subset J

⊂ X is called the

Smarandache Jordan ideal (S-Jordan ideal) if J is a subalgebra of X and for every a
∈ A we have aj and ja ∈ J. We can define Smarandache Jordan left ideal or
Smarandache Jordan right ideal. So Smarandache Jordan ideal is one which is
simultaneously both a S-Jordan left and a S-Jordan right ideal.

D

EFINITION

5.2.6: Let A be a S-Jordan algebra if A has no S- Jordan ideals then we

say A is Smarandache simple Jordan algebra (S-simple Jordan algebra).

D

EFINITION

5.2.7: Let A be a Smarandache Jordan algebra. A pair of distinct

elements x, y

∈ X different from identity of R which are such that xy = yx is said to

be Smarandache pseudo Jordan commutative pair (S-pseudo Jordan commutative
pair) of X, where X is the subalgebra of A which satisfies the Jordan identity if for all
a

∈ A we have x(ay) = (ya)x (or y(ax)) or (xa)y = (ya)x (or (y(ax)). If every

commuting pair of X happens to be S-pseudo Jordan commutative then we say A is a
Smarandache pseudo commutative Jordan algebra (S- pseudo commutative Jordan
algebra).

This concept of pseudo commutative has meaning even if we assume A to be a
commutative algebra. Now we proceed on to define quasi commutativity in non-
commutative Jordan algebras.

D

EFINITION

5.2.8: Let A be non-commutative algebra. A is said to be quasi

commutative if ab = b

a for every pair of elements a, b

∈ A and γ > 1. We say the S-

Jordan algebra A is Smarandache quasi commutative Jordan algebra (S-quasi
commutative Jordan algebra) if every pair of elements in X

⊂ A where X is a

subalgebra, is satisfying the Jordan identity is a quasi commutative algebra.

Thus we see if A is a S-quasi commutative Jordan algebra then A in general need not
be a quasi commutative ring.

D

EFINITION

5.2.9: Let A be an algebra or a Jordan algebra. An element x

∈ A is

said to be semi-nilpotent if x

– x is a nilpotent element of A. If x

– x = 0 we say x is

trivially semi-nilpotent.

T

HEOREM

5.2.7: Let A be a Jordan algebra. Then the following conditions are

equivalent

a. Every element x is semi-nilpotent.
b. Every element x

– x is quasi invertible.

116

Proof: We know if x is semi-nilpotent then x

– x is nilpotent so x

– x is quasi

invertible [37].

If A is only a Smarandache Jordan algebra then we have the modified version of the
above theorem.

T

HEOREM

5.2.8: Let A be a S-Jordan algebra i.e. X

⊂ A satisfies the Jordan identity

then the following are equivalent:

a. Every element x in X is semi-nilpotent.
b. Every element x

– x is quasi invertible.

Proof: Left for the reader to prove.

The concept of semi-nilpotent cannot be defined for Lie algebras as x

= 0 in case of

Lie algebras.

D

EFINITION

5.2.10: We call the algebra A to be reduced if R has no non-zero

nilpotents. We call A to be Smarandache reduced Jordan algebra (S-reduced Jordan
algebra) if the subalgebra X

⊂ A; X satisfies the Jordan identity but X has no non-

zero nilpotents.

Thus we see when A is a S-reduced Jordan algebras A can have non-zero nilpotents
only X

⊂ A mentioned in the definition does not contain non-zero nilpotents.

EFINITION

5.2.11: Let A be a algebra or a S-Jordan algebra, we say A is

Smarandache periodic (S-periodic) if x

= x for some integer n > 1, for all x

∈ X ⊂

A. X is a subalgebra of A which is a Jordan algebra.

Note if n is even in the definition and if A is of characteristic two then we call the
ring A, a Smarandache E-Jordan ring (S-E-Jordan ring).

A quadratic Jordan ring is said to be periodic of every element a in it satisfies the
condition a

= a for some integer n > 1 and has obtained several interesting results in

this direction.

D

EFINITION

5.2.12: Let A be a Jordan algebra or a S-Jordan algebra (X

⊂ A) we say

A is said to be a Smarandache pre J- Jordan algebra (S-pre-J Jordan algebras) if
a

b = ab

for any pair a, b

∈ X ⊂ A and n is a positive integer.

EFINITION

5.2.13: Let A be an algebra which is a S-Jordan algebra we say A is a

Smarandache pre-Boolean Jordan algebra (S-pre-Boolean Jordan algebra) if for X
⊂ A we have xy(x + y) = 0 for x, y ∈ X.

D

EFINITION

5.2.14: Let A be a algebra. We say A is a Smarandache-J-Jordan

algebra (S-J-Jordan algebra) if Y

⊂ X ⊂ A (where X is a subalgebra satisfying the

Jordan identity) if for all y

∈ Y we have y

= y, y > 1.

117

EFINITION

5.2.15: Let A be an algebra

φ ≠ X ⊂ A be a S-Jordan algebra; we say A

satisfies Smarandache Jordan strong ideal property (S-Jordan strong ideal property)
if every pair of ideals of X generates X.

We call A a Smarandache Jordan weak ideal ring (S-Jordan weak ideal ring) if there
exists at least a pair of distinct ideals of X which generate X.

The reader is requested to get results related to these notions.

D

EFINITIONS

5.2.16: Let A and A

be two S-Jordan algebras. We say a map

φ : A →

is said to be a Smarandache Jordan algebra homomorphism (S-Jordan algebra

homomorphism) if

φ from X to X

i.e.

φ : X → X

(where X and X

are subalgebras of

A and A

respectively satisfying the Jordan identity) is a linear transformation as

subspaces. We are not interested whether

φ is defined entirely on A or φ is a linear

transformation from the algebras A to A

EFINITION

5.2.17: Let A be an algebra. We say A is Smarandache power-joined

Jordan algebra (S-power-joined Jordan algebra) if a

= b

for some positive

integers m and n and a, b

∈ X where X ⊂ A is a Jordan algebra.

We say A is Smarandache uniformly power-joined Jordan ring (S-uniformly power-
joined Jordan ring) if m = n i.e. a

= b

for a, b

∈ X.

The reader is expected to obtain some interesting results in this direction.

D

EFINITION

5.2.18: Let A be an S-Jordan algebra, i.e. X

⊂ A be a subalgebra which

satisfies the Jordan identity. We say an element x

∈ A to be Smarandache shrinkable

(S-shrinkable) element if xA = X (Ax = X). The study of the nature of x

∈ A as

interesting for all elements in A cannot serve this purpose; only few special elements
will serve as shrinkable elements. Further every S-Jordan ring need not in general
be shrinkable.

D

EFINITION

5.2.19: Let A be an S-Jordan algebra. Let X

⊂ A be a proper subset of A

which is a subalgebra satisfying the Jordan identity. We say an element m

∈ A, a

Smarandache magnifying element (S-magnifying element) if Xm = A (mX = A). It is
to be noted that all S-Jordan algebras need not have S-magnifying elements.

D

EFINITION

5.2.20: Let A be a S-Jordan algebra we say A is Smarandache dispotent

Jordan algebra (S-dispotent Jordan algebra) if X (

⊂ A) has exactly two S-

idempotents where X is a subalgebra of A satisfying the Jordan identity. If every
subalgebra X of A which satisfies Jordan identity has exactly two S-idempotents then
we call A a Smarandache strong dispotent Jordan algebra (S-strong dispotent
Jordan algebra).

D

EFINITION

5.2.21: Let A be a S- Jordan algebra. We say A is a Smarandache

normal Jordan algebra (S-normal Jordan algebra) if aX = Xa for every a

∈ A where

X is a subalgebra of A which satisfies the Jordan identity.

118

EFINITION

5.2.22: Let A be a S-Jordan algebra. We say A is a Smarandache weakly

G-Jordan ring (S-weakly G-Jordan ring) if for every additive S-semigroup P of X (X
a subalgebra satisfying the Jordan identity) we have xP = Px for every x

∈ X. If xP =

Px = P we say A is a S-G-Jordan ring.

EFINITION

5.2.23: Let A be a S-Jordan algebra. An element x

∈ X ⊂ A is said to be

a Smarandache clean element (S-clean element) if x can be written as a sum of a S-
idempotent and a S-unit in A.

EFINITION

5.2.24: Let A be a S-Jordan algebra. A subset M of X, X

⊂ A is said to

be Smarandache system of local units (S-system of local units) if and only if

a. M consists of commuting S-idempotents.
b. For any x

∈ X and e ∈ M, ex = xe = x.

ROBLEMS

Give examples of S-G-Jordan ring.

Find an example of a S-weakly G-Jordan ring which is not a S-G Jordan
ring.

What is the order of a smallest S-Jordan algebra?

Give an example of a S-strong Jordan algebra?

Find an example of a S-Jordan algebra which has S-shrinkable
elements.

Does every S-Jordan algebra have a S-magnifying element? Justify
your answer.

Illustrate by an example a S-normal Jordan algebra.

Give an example of a S-periodic Jordan ring.

Is the loop ring Z

(2) a S-periodic Jordan ring?

10.

Illustrate by an example a S-pre-Boolean Jordan ring.

11. Is

(3) a S-E-Jordan ring?

12.

Does the ring Z

(3) have semi-nilpotent elements?

119

Chapter 6

SUGGESTIONS FOR FUTURE STUDY

This chapter is unique for it gives suggestions for future study. This chapter is
divided into five sections. First section is devoted to the exposition of groupoid rings
and the probable study of it. As groupoid rings which is the most generalized
structure of non-associative rings till date has not found its place either by way of
research publication or in text form, so the author feels that the study of groupoid
rings will certainly lead to a lot of research. The second section of this chapter gives
a glimpse about the non-associative rings called loop rings.

Research on loop rings is very systematic for there are papers about it but yet there is
not even a single textbook fully devoted to loop rings whereas there are several
books on group rings. In the third section we give the recent concepts about Lie
algebras and Jordan algebras are taken care of in section four. In the final section of
this chapter several other recently discovered non-associative rings found in
literature are given for the reader to make proper Smarandache developments about
them.

6.1 Smarandache groupoid rings

The study of groupoids is very meagre. Except for the book [70] where the notions
of Smarandache groupoids and groupoids are dealt with extensively. So the groupoid
rings which are an enormously a large class of non-associative rings, for they can be
constructed using the infinite number of naturally built groupoids using Z

, Z, Q and

R over any appropriate or needed ring. Except for a few papers on groupoid rings
[31, 64], the study in this direction is totally absent.

Now as groupoid rings are the generalization of group rings, loop rings and
semigroup rings and Smarandache groupoid rings contain always semigroup rings as
subrings. The study of this concept will be beneficial. As Lie identities and Jordan
identities have given rise to two class non-associative algebras, viz., Lie algebras and
Jordan algebras. The first thing the reader can do is to develop using groupoids the
following types of algebras as they are also governed by identities like Moufang,
Bol, etc.

1. Moufang algebras using groupoid rings and excavating the probabilities of

giving them vector space representations. The immediate consequence would
be the study of Smarandache Moufang algebras using groupoid rings.

2. Construction and study of Bol algebras and Smarandache Bol algebras by

using groupoid rings.

Construction and study of P-algebras and Smarandache P-algebras using P-
identities in groupoid rings.

120

Analysis of alternative algebras and Smarandache alternative algebras using
groupoid rings.

As groupoids find its application in Smarandache automaton we can certainly
see that the groupoid rings will find its application in coding theory.

The final thing of importance is that in this book we have shown that groupoid rings,
which are not found to be Jordan algebras, are found to be Smarandache Jordan
algebras. Thus the author at this juncture wants to make known to the researchers
that by studying the Smarandache notions certainly certain subproperties not found
in the total structure can be dealt without complications in a very systematic way by
defining the Smarandache analogue.

Finally the author leaves it to the reader to develop these identities using, the
program first designed and implemented at Clemson University about 1990 by [31].
They have studied groupoids satisfying the identity (xy)z = y(zx) and it has been
analysed.

Thus the reader can use this program in two ways in our groupoids and groupoid
rings:

I. Characterize those groupoid rings satisfying the identities.

II. Using these groupoid to build and develop the theory of formal languages for

which Smarandache groupoids will serve the proper perspective.

6.2 Smarandache loop rings

The study of loop rings has been carried out by several researchers [8 to 16, 21, 24-
29, 49, 50, 57-65]. But the only disadvantage is that we do not have any text solely
on loop rings as we have for group rings or semigroup rings. As loop rings forms the
generalization of group rings and it has more properties in them, as they happen to be
non-associative we felt loop rings should be given a good exposition.

For study of loop rings or Smarandache loop rings the most needed concept is the
loops and Smarandache loops. They have been introduced in [5, 71, 72]. It is still a
pity that we do not have book solely devoted to the study of loops. Hence in this
book we have devoted a complete chapter on the study of loop rings and
Smarandache loop rings.

As Lie algebras or Jordan algebras are algebras which satisfy certain special
identities and are vividly studied, we have several books on them whereas we see
that the loop algebras which are more generalized class of non-associative algebras
does not find its just place in the mathematical world. Alternative algebras has been
intensely studied by [24 to 29] but it could be developed into a book on alternative
and Smarandache alternative algebras.

So we have the following suggestions for the future study.

121

1. Which of the alternative loop rings are Lie or Jordan rings?

2. Construct new class of algebras using loop algebras like "Moufang algebras"

in par with Lie or Jordan algebras. Likewise 'Bol algebras', 'Bruck algebras'
and 'P-algebras' for all these are also algebras satisfying only a special
identity.

It is unknown why such a study has not been carried out in these cases. The author
sees that several of the loop algebras which failed to satisfy Jordan identity happen to
be Smarandache Jordan algebra. So we predict several of the loop rings which does
not satisfy special identities can by all means be Smarandache.

3. Finally the loop rings can be used in the study of algebraic coding theory.

Several of the properties studied for group rings can be studied for Smarandache
loop rings and loop rings. Using the Albert program to study non-associative
algebras satisfying special identities we can find whether the loop rings
/Smarandache loop rings are commutative or associative or neither for special loop
rings satisfying special identities like Moufang, Bol, Bruck or P-identity.

6.3 Lie algebras and Smarandache Lie algebras

There are several excellent books on Lie algebras; but the concept of Smarandache
Lie algebra is introduced only in this book. At the outset we have a lot to develop in
case of Smarandache Lie algebras. As nothing has been done only a very few notions
have been introduced in this book. Second, if one views a Lie algebra as just a non-
associative ring satisfying some special identities then why several of the properties
studied about associative algebras are totally absent in the study of Lie algebras? So
in this book the author has tried to introduce some of the properties found in
associative rings to these algebras.

Further these algebras became little difficult for analysis as x

= 0 for all elements x

in a Lie algebra. Yet we have introduced several concepts in Lie algebras found in
general associative algebras and has left for the reader for further development in this
subject. In all places the Smarandache analogous wherever possible have been
recalled. It will be an ingenious work for any one who takes up solely the study of
Smarandache Lie algebras as there is lot of research to be carried out.

We have introduced Smarandache Lie concepts only for basic Lie algebras so it
remains open for the reader to explore further studies and define new concepts of
Smarandache notions in case of Lie superalgebras graded Lie superalgebras. Further
for the study of complex semi simple Lie algebras refer Springer Monographs in
mathematics 2001 in [47]. The reader has a lot of problem to study about
Smarandache semi simple complex Lie algebras. Also the study of the very recent
textbook for advanced graduate students with all recent Lie algorithms in [19] will
certainly help a researcher to built Smarandache Lie algebras in a very systematic
and in an innovative way.

122

As there are several text books on Lie algebras, monographs and collected papers
which can give an excellent exposition of Lie algebras we felt it would be sufficient
to give a simple exposition about Smarandache Lie algebras and some of its basic
properties and thus by leaving a greater scope for the reader to develop nice books on
Smarandache Lie algebra.

Also it has become inevitable to mention here that for a solid development of
Smarandache Lie algebra one should at least develop four Smarandache concepts
viz. Smarandache Homological algebra, Smarandache manifolds, Smarandache
topology and Smarandache linear algebras. Finally the dictionary on Lie algebras
severs the goal of desktop reference [23]. For any beginner in Lie algebra as well as
for any researcher this will serve as a boon Thus the reader has lot for his future
study on the concept of Smarandache Lie algebras.

6.4 Smarandache Jordan Algebras

The study of Jordan algebras started as early as 1940s. A complete and an excellent
exposition of Jordan algebras was given in 1968 by the famous algebraist N.
Jacobson in his book entitled "Structure and representations of Jordan algebras".
Till now a lot of research has been carried out and several others have written lecture
notes and books on Jordan algebras. So we have not taken up the Smarandache
Jordan algebras in an exhaustive way. But we are happy to state that we have
introduced several notions which are common to associative algebras and analysed it
in the case of Jordan algebras. We have introduced several Smarandache Jordan
concepts but we have not developed them into big theories as we felt that any
interested reader can develop them as it can be done as a matter of routine.

Further any researcher has always the flexibility to define Smarandache Jordan
algebra in a different ways. We at the outset call any non-associative algebra A to be
a Smarandache Jordan algebra if it has a proper subset X which is a subalgebra and
satisfies the Jordan identity. We can always build non-trivial Smarandache Jordan
algebras by using Smarandache mixed direct product of algebras in which we insist
that one of the algebras must be a Jordan algebra so that we have a class of
Smarandache Jordan algebras.

It is pertinent to mention here that by the Smarandache mixed direct product of
algebras we can simultaneously have a algebra SA

to be both a Smarandache

Jordan algebra and Smarandache Lie algebra and so on i.e. by defining Smarandache
mixed direct product of type M as follows:

Let SA

= L

× … × L

be the mixed direct product of non-associative algebras L

…, L

defined over the same field K where at least one of the L

's is a Jordan algebra

and one of the L

's is a Lie algebra then the Smarandache mixed direct product of

algebras of type M is a Smarandache Jordan algebra and Smarandache Lie algebra
simultaneously. Except for the Smarandache mixed direct product we will not be in a
position to see both structure embedded in a single algebra. If in the S-mixed direct
product of type M we take one of the L

's be a Smarandache loop algebra over the

field K then SA

will also be a Smarandache loop algebra. Thus study of such

123

Smarandache mixed direct products will certainly give several nice and interesting
theories.

Study of Smarandache Jordan superalgebras have not been carried out by the author.
The author suggests the reader to refer the Lecture notes 211 [18] for the study of
Jordan superalgebras so that they can develop Smarandache Jordan superalgebras.
One of the certain ways of developing is by including in the Smarandache mixed
direct product of type M a L

to be a Jordan superalgebra i.e. we will call a non-

associative algebra A to be a Smarandache Jordan superalgebra if A has a proper
subset X such that X is a Jordan superalgebra.

Thus we have the algebra SA

to be simultaneously a Smarandache Jordan algebra

and Smarandache Jordan superalgebra if in the product of L

's we have a L

and a L

distinct to be Jordan algebra and a Jordan superalgebra respectively.

On similar lines by taking in the Smarandache mixed direct product of algebras of
type M i.e. in SA

one of the L

's to be a Lie superalgebra we can say SA

is a

Smarandache Lie superalgebra. Thus we want to emphasize that these Smarandache
mixed direct product of non-associative algebras of type M can give an algebra say
SA

which can be at a time a Smarandache Jordan algebra, Smarandache Lie

algebra, Smarandache Lie superalgebra, Smarandache Jordan superalgebra and so
on. But otherwise getting a same algebra to have all such properties is impossible.

Thus the reader using these concepts and notions is advised to develop a book of
research on Smarandache Jordan algebras.

6.5 Other non-associative rings and algebras and their Smarandache
analogues

This section mainly introduces the concepts of non-associative algebras like
Bernstein, Sagle algebra, finite dimensional complex filiform Leibniz algebra,
composition algebra, Akisvis algebra and their Smarandache analogue. Thus the
researcher has a lot of scope to develop these ideas. One of the recent study in non-
associative algebras is the notion of Bernstein algebras

D

EFINITION

[18]: A basic algebra A over a field F, char F

≠ 2 is a non-associative

commutative algebra with a non-zero homomorphism w : A

→ F (weight

homomorphism). Among basic algebras, those that satisfy the identity (x

)

= w(x)

for all x

∈ A are called Bernstein algebras.

It is well known that Bernstein algebra A always has idempotent elements and if e

e is one of them then w(e) = 1 and ker w = U

⊕ V where U = {u ∈ ker w / 2eu = u}

and V = {v

∈ ker w / ev = 0}.

The set of idempotent elements of the Bernstein algebra A given by I(A) = {e + u +
u

/ u

∈ U}.

We define Smarandache Bernstein algebra as follows.

124

EFINITION

6.5.1: A basic algebra A over a field F, char F

≠ 2 is a non-associative

algebra with a non-zero homomorphism w : A

→ F such that the restriction of w

from B

⊂ A where B ≠ φ (B is a commutative algebra) say w

: B

→ F such that (x

)

= w

(x)

for all x

∈ B. Then we call A the Smarandache Bernstein algebra (S-

Bernstein algebra).

Several of the properties enjoyed by the Bernstein algebra can be defined and studied
in case of Smarandache Bernstein algebras.

D

EFINITION

6.5.2: An algebra satisfying the identities x

= 0, J(x, y, z)w = (J(w, z,

xy) + J(w, y, zx) + J(w, x, yz) is called a Sagle algebra where J(x, y, z) = (xy)z +
(zx)y + (yz)x.

We define Smarandache Sagle algebras as follows:

D

EFINITION

6.5.3: Let A be any algebra. We call A a Smarandache Sagle algebra (S-

Sagle algebras) if A has a proper subset B, (B

⊂ A, B ≠ φ) such that B is a Sagle

algebra.

T

HEOREM

6.5.1: If A is a Sagle algebra having proper subalgebras then A is a S-

Sagle algebra.

Proof: Direct by the very definition.

It is left as an exercise for the reader to prove the following theorem:

T

HEOREM

6.5.2: Let L be a S-Sagle algebra then L need not in general be a Sagle

algebra.

For Sagle algebras the reader can develop the concepts analogous to any ring
theoretical properties.

Here we describe the finite dimensional complex fillform Leibniz algebras as right
multiplication algebras that satisfy abelianness condition. These algebras are
generalizations of Lie algebras and are defined by the identity [x[y, z] = [[x, y], z] –
[[x, z], y] where [x, y] denotes the algebra product.

If L is a Leibniz algebra the series C

(L) = [L, L] and C

k+1

(L) = [C

(L), L] is defined

inductively. The algebra L is called nilpotent if C

(L) = 0 for some k, if a stronger

condition dim(C

(L)) = dim(L) – k – 1 holds for k = 1, 2, …, dim L – 1, then L is

called filiform.

The purpose of this paper is to classify those filiform algebra that satisfy an
additional condition namely [L, L] is contained in the centre of L.

D

EFINITION

6.5.4: We define Smarandache finite dimensional filiform Leibniz

algebras (S-finite dimensional filiform Leibniz algebras) as algebras which has a
subalgebra which is finite dimensional filiform Leibniz algebras.

125

The easiest way of constructing them is by using S-mixed direct product. If we take
in the S-mixed direct product type M i.e. in SA

an algebra L

to be a filiform

Leibniz algebra then the resulting SA

will be a Smarandache filiform Leibniz

algebra.

A non-associative algebra C over a field F is called a composition algebra if there
exists a quadratic form q : C

→ F such that q(xy) = q(x)q(y) for all x, y ∈C whose

induced bilinear form b

(x, y) = q(x + y) – q(x) – q(y) is non-degenerate.

Composition algebras containing a unity element are well-known. Here symmetric
composition algebras are investigated. These satisfy additional equation b

(xy, z) =

(x, yz)

∀ x, y, z ∈ C which forces the dimension of C to be either 1, 4, 2, 8.

EFINITION

6.5.5: Smarandache composition algebras (S-composition algebras) are

algebras A which has a subalgebra X

⊂ A such that X is a composition algebra.

Smarandache mixed direct product SA

easily yields such S-composition algebras.

EFINITION

[48]: An Akivis algebra is a vector space V endowed with a skew

symmetric bilinear product (x, y) a [x, y] : A × A → A and a trilinear product (x, y,

→ 〈x, y, z〉: A × A × A → A such that these two operations are related by the so-

called Akivis identity which is a generalized form of the Jacobi identity.

From any non-associative algebra (B, ., +) one obtains a Akivis algebra A(B) by
putting [x, y] = x . y – y . x and

〈x, y, z〉 = (x . y) . z – x . (y . z).

We define Smarandache Akivis algebras as follows:

D

EFINITION

6.5.6: Let V be a Smarandache vector space. If a subalgebra X of V

satisfies the Akivis identity then we call V a Smarandache Akivis algebra (S-Akivis
algebra).

T

HEOREM

6.5.3: If V is a Akivis algebra which has a S-vector space then V is a S-

Akivis algebra.

Proof: Follows directly from the very definitions.

The reader can develop all Smarandache concepts in Akivis algebras and study them
as further research on these structures. Thus we see that by introduction of several
new Smarandache concepts in these non-associative structures any researcher can
always develop not only the existing theories but also new Smarandache theories.

For the study of these structures the available references has been provided but we
don't claim that we have exhausted all the references. Only what was available to us
has been provided.

Finally these will find its main application in algebraic coding theory, which is a
very fast developing area of research. So when Smarandache concepts are introduced
it will be much more helpful. Thus this book totally provides immense scope for the
reader to develop the subject.

126

A few simple problems are suggested for the reader.

P

ROBLEMS

1.  Give an example of a Akisvis algebra.
2.  Does there exist a Akisvis algebra which is not a S-Akisvis algebra.
3.  Give an example of a S-Sagle algebra.
4.  Does there exist a finite dimensional Bernstein algebra?

127

Chapter 7

SUGGESTED PROBLEMS

This chapter suggests 150 problems on non-associative rings and related concepts.
Certainly these problems will not only make the researcher get more insight into the
subject but also help him to get fixed ideas by tackling these problems. Some of
these problems can even be very challenging research problems. Whereas problems
of introducing and characterizing Moufang algebras, Bol algebras or Bruck algebras
will give us bulk results, as such this type of trial to the best of my knowledge has
not been carried out by anyone. Further some of the solutions to these problems can
generate substantial material for research. As understanding of any subject lies in the
ability to solve related problems, we have suggested the following problems. We
wish to state that we have not included all the problems, which we have mentioned
in our suggestions for future study.

1. Let

and

be two SNA-rings.

φ : R to R

be a S-homomorphism find

conditions on

φ so that Ker φ is an ideal of R. (Hint: We don’t assume Ker φ =

(0), we assume Ker

φ ≠ (0)).

Find the SNA-ring of least order.

Is the smallest non-associative ring a SNA-ring?

Can on Z

be defined two binary operations so that Z

is a non-associative ring?

Find the smallest SNA-ring, which is a SNA-Bol ring.

Does there exist SNA-rings other than the ones got from

a. loop rings?
b. Groupoid rings?

Find a SNA-ring in which every ideal is a SNA-ideal of R.

Find conditions on a SNA-ring R so that every subring of R is a SNA-subring of
R.

a. Characterize SNA-rings which has ideals none of which are SNA-

ideals

b. Characterize SNA-rings which has subrings none of which are SNA-

rings.

Characterize those SNA-rings which are SNA-simple.

128

10.

Let R be SNA-ring having a maximal SNA-ideal say I relative to a SNA-subring
B of R. Find conditions on R and B so that R/I is simple.

11.

For the loop ring Z

(m) n > 3, n odd and m

≤ n, (m, n) = 1 = (m, n-1) (for a

fixed n)

a. How many S-right ideal exists in Z

(m)?

b. What is the lattice structure of S-right ideals?
c. How many right ideals exists in Z

(m)?

d. What is the lattice structure of right ideals?
e. Compare the lattice structures in (b) and (d).

12.

Study the same problem for

a. left

ideals.

b. SNA-left ideals.
c. ideals and
d. SNA-ideals.

13. Let

(m) be the loop ring. Study problems (11) and (12) when

a.  (t, n + 1) = 1
b.  (t, n + 1) = t; t a prime
c.  (t, n) = 1
d.  (t, n) = t
e.  (t, n) = d
f.  (t, n + 1) = d

14.

Characterize those units in ZL

(m); L

(m)

∈ L

which are S-NA units.

15.

Find all units in Z

(m); L

(m)

∈ L

a. Which are not SNA-units
b. Which are SNA-units.

16. Find conditions on n and p so that the loop ring Z

(m) has S-idempotents

where (n, p) = 1 or (n, p) = d.

17.

Find necessary and sufficient conditions for the loop ring Z

(m) where L

(m)

∈ L

to have nontrivial S-quasi regular elements.

18. Does

(m) have S-quasi regular elements x where |supp x| is an even number?

19. Find

S(J(Z

(m)).

20.

Is SJ (Z

(m))

≠ φ, if (n, p) = 1? Justify your answer.

21.

Find SJ (ZL

(m)).

129

22.

Characterize those NA rings in which every quasi regular element is a S-quasi
regular element.

23.

Can the loops in the class L

for any suitable ring or field contribute quasi regular

elements which are also S-quasi regular elements?

24.

Characterize those loop rings or NA rings in which J(R) = SJ(R).

25.

Find a necessary and sufficient condition for any NA-ring to be both strictly right
ring and S-strictly right ring.

26.

Find condition for the loop ring Z

(m) to be

a. strictly right ring.
b. S-strictly right ring.

27.

Classify those NA-rings which satisfy only strictly right ring property and does
not satisfy the S-strictly right ring property.

Note: Study the same problem just in case of ideals.

28.

Give an example of a S-ring which is not a S-strictly right ring.

29.

Determine all S-semi-idempotents of ZL

(m); L

(m)

∈ L

30. Find conditions on n and t so that the collection of S-semi-idempotents in Z

(m) is non-empty.

31.

When will SI (R) = SSI(R)? Find conditions on the NA ring R.

32. Can

(m) when L

(m) is a non-commutative loop be a S-Marot ring? Justify

your answer. Characterize those loops in L

which are not commutative still they

happen to be

a. S-weak Marot ring.
b. S-Marot ring.

33.

Let L be any loop, K any field and KL the loop ring of the loop L over K. Find
all idempotents in L

∗

, can KL have idempotents which are Smarandache

idempotents? Find a necessary and sufficient condition on loop ring KL, so that

a. L

∗

is a loop

b. L

∗

is only a groupoid

Can L

∗

every be a semigroup? Justify your claim.

34.

Characterize those loops L so that L

∗

= S (L

∗

35. Let

(m)

∈ L

and K be any field. Find conditions on KL

(m) so that

130

a. S ((L

(m))

∗

)

≠ 0

b. Find the number of elements in S ((L

(m))

∗

) or equivalently find the

number of S-subloops in KL

(m).

c. When will S (L

(m))

∗

) be a loop?

d. When will S ((L

(m))

∗

) be only a groupoid?

36.

In a loop L, is the concept of t.u.p loops and u.p loop equivalent or distinct?

37.

Let L be a loop which is not a u.p loop. Can L be a S.u.p loop? Find conditions
on L so that L is a S-strong u.p loop but is not a u.p loop.

38.

Find those loops in L

(m)

∈ L

which are

a. S-u.p

loops.

b. S-S.u.p loops.

39.

Give an example of a Smarandache S.u.p loop which is not a u.p loop.

40.

Can we say if L is a u.p loop or a t.u.p loop, L can be S.u.p loop, or S.S. u.p loop
or S.t.u.p loop or S.S.t.u.p loop? (Hint: Illustrate by examples. This question is
vital as we may have S-subloops A of L. It may happen L is a u.p loop still A

⊂

L will not be a u.p loop).

41.

If L is a S.S.u.p loop or a S.S.t.u.p loop will L be a u.p loop or a t.u.p loop?

42.

Which of the loops in L

(m) are S-Hamiltonian loops?

43.

Characterize those loops of odd order which are S-Hamiltonian.

44. Let

be a ring and L be a loop of order m;

a. If (n, m) = 1 when does the loop ring Z

L contain nontrivial

orthogonal ideals?

b. If (m, n) = d, can we say Z

L contains always a pair of nontrivial

orthogonal ideals?

45.

Obtain a necessary and sufficient condition on the loop ring FL so that every pair
of ideals in FL is orthogonal.

46.

Characterize those loop rings in which every pair of orthogonal ideals is also a S-
orthogonal ideal.

47.

Characterize those loop rings which are

a.  Strictly right commutative.
b.  Strongly right commutative.
c.  Right commutative.

48. Let

(m) be the loop ring L

(m)

∈ L

. Can the loop ring Z

131

a. S-strongly right commutative
b. S-right commutative.

49.

Characterize those loop rings which are inner commutative.

50.

Which of the loop rings ZL

(m) are inner commutative? Characterize those loop

rings which are inner commutative but not S-inner commutative.

51. Characterize those loop rings which are inner commutative and S-inner

commutative.

52.

Obtain a necessary and sufficient condition for the loop ring RL to have only one
S(P(RL)). (Hint: It is assumed that RL has more than one S-subring).

53.

Characterize those loop rings RL which are such that each S-subring A of L give
a distinct S(P(RL)).

54.

Characterize those loop rings RL in which P(RL) = S(P(RL)).

55.

Characterize those loop rings in which all ideals and S-ideals coincide.

56.

Characterize those loop rings in which the S-ideals form a

a. Super modular lattice.
b. Strongly modular lattice.

57. Let

(m)

∈ L

and Z the ring of integers Find the lattice structure of

a. ideals of ZL

(m).

b. S-ideals of ZL

(m).

c. S-subrings of ZL

(m) and

d. Subrings of ZL

(m).

Study problem (57) in which Z is replaced by Z

58.

Find conditions on the loop ring ZL

(m) or Z

(m) so that the

a. SN

(ZL

(m)) = N (ZL

(m)).

b. SN (Z

(m)) = N (Z

(m)).

c. Z

(ZL

(m) = S (Z (ZL

(m)).

d. Z (Z

(m)) = SZ (Z

(m)).

59.

Does the center and nucleus of RL

(m) coincide?

60.

Does the S-center and S-nucleus coincide on RL

(m)?

61.

Find all units and S-units in the loop ring ZL

(m). Do we have loop ring so that

both the sets are identical?

62.

Find conditions on the loop ring RL so that S-units in RL are central units.

132

63.

Characterize those vector spaces V which has S-basis which are S-strong basis of
V.

64. Characterize those vector spaces V for which every basis is a S-basis. Also

characterize those vector spaces which has no S-basis.

65.

Characterize those vector spaces which has S-basis.

66.

Can we have any vector space V in which every basis is a S-basis?

67.

Study Zassenhaus conjecture for the class of loop rings ZL

(m).

68.

Verify S-Zassenhaus conjecture for the new class of loop rings ZL

(m).

69.

Find the S-normalized units of ZL

(m).

70.

Find S U (ZL

(m)).

71. Find

V(ZL

(m)).

72. Find all groupoid rings RG (necessary and sufficient condition) so that every

idempotent in RG is a S-idempotent of RG. Another extreme case is no
idempotent in RG is a S-idempotent of RG.

73.

Find whether the groupoid rings ZG where G is a groupoid built using modulo
integers be a

a. n-ideal

ring.

b. S-n-ideal ring.

74.

Study the above question when Z is replaced by Z

, where Z

is:

a. Any prime field Z

b. Any ring Z

75.

Characterize or prove or disprove the existence of groupoid rings in which the
notion of S-n-ideal ring and n-ideal ring coincides.

76. Let

G be a groupoid ring. Find conditions on G and Z

so that the groupoid

ring Z

G has every generalized semi-ideal to be a S-generalized semi-ideal of

77. Find necessary and sufficient condition for the groupoid ring RG to have no

generalized semi-ideal to be a S-generalized semi-ideal of RG.

78. Find the number of S-mod p envelope SG

∗

for the groupoids in Z(n), Z

∗

(n),

∗∗

(n) and Z

∗∗∗

(n), where n is a fixed number.

133

79.

Prove or disprove the number of elements in SG

∗

depends also on the field Z

choose over which the groupoid rings are defined.

80.

Does there exist a groupoid G which has only one S-subgroupoid such that SG

∗

? Justify your claim.

81.

Classify those groupoid rings using the groupoids from Z(n), Z

∗

(n), Z

∗∗

(n) and

∗∗∗

(n) in which for suitable rings

a.  The minimal and maximal ideal coincide.
b.  The S- maximal ideals
c.  The S- minimal ideals.
d.  Can S-maximal ideals and S-minimal ideals be coincident in ZG

where G

∈ Z

∗∗∗

(n)?

82.

Does the groupoid ring Z

G where G

∈ Z

∗∗∗

(n) have the set of all ideals to form a

modular lattice?

83.

Study the problem 82 in case of

a. S-ideals
b. S-pseudo ideals

84.

Find necessary and sufficient conditions so that the groupoid ring KG satisfies

a. A.C.C
b. D.C.C.

85.

Find a necessary and sufficient conditions for the groupoid rings RG to satisfy

a. S-A.C.C
b. S-D.C.C

on S-ideals of RG.

86.

Characterize those groupoid rings which are S-E-rings.

87.

Obtain a necessary and sufficient condition for the groupoid ring to be a S-p-ring.

88.

Find a class of groupoid rings which are S-pre-J-rings.

89.

Show that the class of groupoid rings which are S-reduced is nonempty.

90.

Determine a class of S-quasi commutative groupoid rings.

91.

Characterize those groupoid rings which are Lin rings.

92.

Characterize those groupoid rings RG in which every ideal is a S-ideal. Hence or
otherwise, prove that RG is both a ideally obedient ring and S-ideally obedient
ring.

134

93.

Characterize the class of groupoid rings which are S-I

∗- rings.

94.

Study problem (93) for the groupoid ring RG where G

∈ Z

∗∗∗

(n).

95.

Find necessary and sufficient condition for the groupoid rings Z

G where G

∈

∗∗∗

(n) to be S-quotient-ring (S-Q ring) with Z

G where G

∈ Z

∗∗∗

(r).

96.

Characterize those groupoid rings RG (where G

∈ Z

∗∗∗

(n)) so that

a. RG is never a F-ring.
b. RG is always a F-ring.

97.

Find conditions on G and on the ring R so that the groupoid ring RG is a

-ring.

98.

For the groupoid L = { [-

∝, ∝], ∗, (m , n)} find associator and commutator.

99.

Using the groupoid L given in the above problem. Study the groupoid ring Z

ZL and any RL.

100. Find a necessary and sufficient condition for a groupoid ring RG to be a S-

normal groupoid ring.

101.

Does the new class of groupoids L = {[-

∝, ∝], ∗ , (m , n)} and their groupoid

rings enjoy any new property if

a.  (m, n) = 1.
b.  if n = m – 1 or m = (n – 1).
c.  if m and n are both primes.

102.

Obtain a necessary and sufficient condition on the groupoid ring RG to satisfy
the identity (xy)x = x(yx) for all x, y

∈ RG.

103.

Is it possible to find S-exponentiation ring using groupoids without unit? Or if
the groupoids G are replaced by S-groupoid will we have RG to be a S-
exponentiation ring? Justify your answer.

104.

Classify those groupoid rings which has n-capacitor subgroups.

105. Characterize those commutative groupoid rings which do not have n-capacitor

subgroups.

106.

Characterize those groupoid rings RG which are

a. essential.
b. S- essential.

for R = Z

and G

∈ Z

∗∗

(n).

135

107.

Find the class of groupoid rings Z

G where G

∈ Z

∗∗∗

(n) which are

a. not

essential

b. not S-essential

108.

Characterize those groupoid rings Z

G which has

a. non trivial radical.
b. non trivial S-radical.

where

∈ Z

∗∗∗

(n) for

a.  (n , p) = 1.
b.  (n , p) = p.
c.  n = p.

109.

Find a necessary and sufficient condition on the groupoid rings Z

G where m is a

prime or a composite number and G

∈ Z

∗∗∗

(n) to be S-Lie ring.

110.

Find necessary and sufficient condition on the loop ring Z

(m) to a S-Lie ring

where K may be a prime or a composite number and L

(m)

∈ L

111. The problems (109) and (110) can be reformulated as follows: Let Z

G or

(m)

be a groupoid ring of a loop ring. Find a necessary and sufficient

condition for the subring U =

{

}

that

such

)

(

∑

∈

∑

to be

a. Smarandache NA ring
b. Satisfy the identities x

= 0 and (xy)z + (yz)x + (zx)y = 0 i.e. U

satisfies Lie identities. Is Z

G or Z

(m) is a S-Lie algebra?

112.

Characterize those groupoid rings Z

G where G

∈ Z

∗∗∗

(n) such that U = {

α =

Σα

∈ Z

G /

Σα

= 0} is a SNA ring.

113.

Study the same problem 112 in the context of loop rings.

114.

Characterize those S-Lie algebras which have S-Lie maximal ideals which are
not maximal ideals in general.

115.

Does there exist examples of S-Lie minimal ideals which are not minimal ideals.
Justify your answer.

116.

Describe those S-Lie algebras

a. which has no S-pseudo Lie ideals
b. which has S-pseudo Lie ideals.

117.

Is the S-pseudo quotient ring L/X a S-Lie ring?

136

118.

What is the structure of L/X described in problem 117? Is it at least a Lie ring?
Justify your answers.

119.

Characterize those S-Lie rings which are S-filial Lie rings.

120.

Let L be a Lie ring. Find necessary and sufficient conditions for L to be a S-filial
Lie ring.

121.

Characterize those Lie rings which are

a. S-Lie subsemi-ideal ring.
b. Subsemi ideal ring.

122.

Characterize those Lie rings which are

a. S-semiprime.
b. Semiprime.

123.

Characterize those Lie rings which are

a. S-Chinese Lie rings.
b. Chinese Lie rings.

124.

Characterize those Lie rings which are never Chinese Lie rings or S-Chinese Lie
rings.

125.

Characterize those Lie algebras which are

a. Lie multiplication ideals
b. S-Lie multiplication ideals

126.

Characterize those Lie algebras which has no Lie multiplication ideals.

127.

Study those Lie algebras which has S-Lie multiplication ideals but has no Lie
multiplication ideals.

128.

Characterize those Lie algebras which satisfy

a. Super ore condition.
b. S-super ore condition.

129.

Describe those Lie algebras which are

a. Obedient Lie rings.
b. S-obedient Lie rings.

130.

Characterize those Lie algebras in which every subalgebra of same dimension are

a. Conjugate.
b. S-conjugate.

137

131.

Characterize those Lie algebras which are generalized stable algebras.

132.

Classify those Lie algebras which are S-generalized stable Lie algebras.

133.

Obtain a necessary and sufficient condition for Lie algebras to be

a. n-ring

ideal.

b. S-n-ring ideal.

134.

Derive a necessary and sufficient condition for a Lie algebra to be Smarandache
semi-simple.

135.

Characterize those Lie algebras which are S-simple.

136.

Obtain some interesting results on

a. S-Lie simple algebras.
b. S-Lie semi-simple algebras.

137.

Characterize those Lie algebras which are S-strongly Lie simple.

138.

Characterize those Lie algebras which are S-strongly Lie semi-simple.

139.

Does there exist any relations or marked differences between S-strongly Lie
simple algebras and S-strongly Lie semi-simple algebra? Justify your answer
with explicit examples.

140.

Can we prove any S-ideal is a S-semi-simple Lie algebra of characteristic 0 is S-
semi-simple.

141.

Classify those Lie algebras which has S-Cartan subalgebras.

142.

Characterize those Lie algebras which are both S-Lie simple and S-Lie semi-
simple.

143.

Obtain a necessary and sufficient condition so that every S-Jordan algebra has a
S-magnifying element.

144.

Characterize those S-Jordan algebras which are S-shrinkable.

145.

Find a necessary and sufficient condition for a S-Jordan algebra to be S-dispotent
Jordan algebra.

146.

Obtain a necessary and sufficient condition for a S-Jordan ring to be

a. S-weakly G-Jordan ring.
b. S-G-Jordan rings.

147.

Characterize those S-Jordan rings which has at least one S-clean elements.

138

148.

Can there be a S-Jordan ring in which every element can be a S-clean element?

149.

Characterize those S-Jordan rings which has no S-clean elements.

150.

Characterize those Akisvis algebras which are S-Akisvis algebras.

139

REFERENCES

1. A

LBERT

, A.A., Non-associative algebra I, II, Ann. Math.(2), 43, 685-707,

(1942).

2. A

LBERT

, A.A., A structure theory for Jordan algebras, Ann. Math(2), 48,

546-567 (1947).

3. B

ERNAVIDES

, R., B

URGUENO

and

OSTA

, Mallol R., Some algebras of order

four, Int. J. Math. Game Theory and Algebra, 9, 159-172 (1999).

4. B

RUCK

.R.H., Some results in the theory of linear non-associative algebra,

Trans. Amer. Math. Soc., 56, 141-199 (1944).

5. B

RUCK

.R.H., A survey of binary systems, Springer-Verlag, (1958).

6. C

ASTILLO

J., The Smarandache semigroup, International Conference on

Combinatorial Methods in Mathematics, II meeting of the project 'Algebra,
Geometria e Combinatoria', Faculdade de Ciencias da Universidade do Porto,
Portugal, 9-11 July 1998.

7. C

HANG QUAN

, Zhang, Inner commutative rings, Sictiuan Dascue Xuebao

(Special issue), 26, 95-97 (1989).

8. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Isomorphism of loops which have

alternative loop rings, Comm. Algebra, 13, 1-20 (1985).

9. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Loops whose Loop rings are

alternative, Comm. Algebra, 14, 293-310 (1986).

10. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Is a right alternative loop ring

alternative?, Algebras, Groups and Geometries, 5, 297-304 (1988).

11. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Code loops are RA2- loops, J.

Algebra, 130, 385-387 (1990).

12. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Loops whose loop rings are

alternative, Comm. Algebra, 18, 659-668 (1990).

13. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Loops whose loop rings in

characteristic two are alternative, Comm. Algebra, 18, 659-688 (1990).

14. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Moufang loops with a unique non-

identity commutators, J. Algebra, 130, 369-384 (1990).

140

15. C

HEIN

, Orin and G

OODAIRE

, Edgar G., Minimally non-associative Moufang

loops with a unique non-identity commutator are ring alternative, Comment
Math. Univ. Carolinae, 43, 1-8 (2002).

16. C

HEIN

, Orin, A short note on super nuclear (central) elements of inverse

property loops, Archiv der Mathematik, 33, 131-132 (1979).

17. C

ONNELL

, I.G., On the group ring, Canadian J. Math., 15, 650-667 (1936).

18. C

OSTA

, Roberto, G

RISHKOV

, Alexander, G

UZZO

Jr., Henrique and P

ERESI

, A.

Luiz, Non-associative algebras and its applications, Lecture notes in Pure
and Applied Mathematics, 211, Marcel Dekker, (2000).

19. D

RAAF

, Willem A., Lie Algebras :Theory and Algorithms, North-Holland

Pub. Co., Amsterdam, (2000).

20. D

UBROVIN

, I.N., Chain rings, Rad. Math., 5, 97-106 (1989).

21. E

LDUQUE

, Alberto and H

, C. Myung, Mutations of alternative algebras,

Kluwer Acad. Pub., Boston, (1994).

22. E

LDUQUE

, Alberto, A note on triality and symmetric composition, Abh. Math

Sem. Univ, Hamburg, 70, 125-135 (2000).

23. F

RAPPAT

, L., S

CIARRINO

, A., and

ORBA

, P., Dictionary on Lie algebras and

superalgebras, Acad. Press, (2000).

24. G

OODAIRE

, Edgar G. and M

ILES

, Cesar Polcino, Isomorphism of integral

alternative loop rings, Rendiconti Del Circolo Matematico Di Palermo, 37,
126-135 (1988).

25. G

OODAIRE

, Edgar G. and M

ILES

, Cesar Polcino, Torsion units in alternative

loop rings, Proc. Amer. Math. Soc., 107, 7-15 (1989).

26. G

OODAIRE

, Edgar G. and P

ARMENTER

, M.M., Units in Alternative loop rings,

Israel J. Math., 53, 209-216 (1986).

27. G

OODAIRE

, Edgar G. and P

ARMENTER

, M.M., Semi-simplicity of alternative

loop rings, Acta. Math. Hung, 50, 241-247 (1987).

28. G

OODAIRE

, Edgar G., Alternative loop rings, Publications Mathematical

Tomus, 30, 31-38 (1983).

29. G

OODAIRE

, Edgar G., Groups embeddable in alternative loop rings, Contrib.

General Algebra, Verlag, 169-176 (1991).

141

30. G

RAY

, Mary W., A radical approach to algebra, Addison-Wesley, (1970).

31. H

ENTZEL

, Irvin Roy, J

ACOBS

, D.P., and M

UDDDANA

, Sekhar V.,

Experimenting with the identity (xy)z = y(zx), (1993)
http://www.cs.clemson.edu/~dpj/albertstuff/albertpapers/xyz.ps

32. H

IGMAN

, G., The units of group rings, Proc. London Math. Soc., 46, 231-248

(1940).

33. I

QBAL

NNISA

and V

ASANTHA

ANDASAMY

, W. B., Supermodular lattices, J.

Madras Univ., 44, 58-80 (1981).

34. J

ACOBSON

, F.D., and J

ACOBSON

, Nathan, Classification and representation of

semi simple Jordan Algebras, Trans. Amer. Math. Soc., 65, 141-169 (1949).

35. J

ACOBSON

, Nathan, General representation theory of Jordan algebras, Trans.

Amer. Math. Soc., 70, 509-530 (1951).

36. J

ACOBSON

, Nathan, Lie algebras, Interscience Pub. Wiley & Sons, (1962).

37. J

ACOBSON

, Nathan, Structure and representations of Jordan algebras, Amer.

Math. Soc. Colloq Pub., (1968).

38. J

OHNSON

, P.L., The modular group ring of a finite p-group, Proc. Amer.

Math. Soc., 68, 19-22 (1978).

39. K

OECHER

, Max, The Minnesota notes on Jordan algebras and their

applications, Lecture Notes in Matematics, 1710, Springer Verlag, (1999).

40. M

ALLOL

, Cristian and V

ARRO

, Richard, Automorphism and derivations in

Jordan algebras of rank 3, Algebras, Groups and Geometries, 14, 297-311
(1997). [MR: 2000b, 17042].

41. M

ATSUDA

, R.M., On Marot rings, Proc. Japan Acad., 60, 134-138 (1984).

42. N

ORTON

, D.A., Hamiltonian loop, Proc. Amer. Math. Soc., 3, 56-65 (1952).

43. P

ADILLA

, Raul, Smarandache algebraic structures, Bull. Pure Appl. Sci.,

17E, 199-121 (1998).

44. P

ASSMAN

D.S., The algebraic structure of group rings, Wiley Interscience,

(1977).

45. R

, Urmie, A characterization theorem for a certain class of graded Lie

superalgebras, J. Algebra, 229, 405-434 (2000).

142

46. R

EDFIELD

, R.H., Exponentiation in rings, Amer. Math. Monthly, 102, 36-40

(1995).

47. S

ERRE

, Jean-Pierre, Complex semi-simple Lie algebras, translated from the

French by G.A. Jones, Reprint of the 1987 edition, Springer Monographs in
Mahematics, Springer Verlag, Berlin, (2001).

48. S

HESTAKOV

, Ivan P., Every Akivis algebra is linear, Geom. Dedicata, 77,

215-223 (1999).

49. S

INGH

.S.V., On a new class of loops and loop rings, PhD thesis (under the

guidance of W.B.Vasantha) Indian Institute of Technology, Aug. 1994.

50. S

OLARIN

, A.R.T, A short note on Bol algebras, Riv. Mat. Pure. Appl., 8, 7-11

(1988).

51. S

PRINGER

, T.A., Jordan algebras and algebraic groups, Springer-Verlag,

(1973).

52. S

TROJNOWSKI

, A., A note on unique product groups, Comm. Algebra, 8, 235-

260 (1980).

53. V

ASANTHA

ANDASAMY

, W. B., Studies in algebra with special reference to

simple modular lattices and supermodular lattices, Ph.D. Thesis, Univ. of
Madras, (1980).

54. V

ASANTHA

ANDASAMY

, W. B., On zero divisors in reduced group rings

over ordered rings, Proc. Japan Acad., 60, No.9, 333-334 (1984).

55. V

ASANTHA

ANDASAMY

, W. B., A note on the modular group ring of a finite

p-group, Kyungpook Math. J., 26, 163-166 (1986).

56. V

ASANTHA

ANDASAMY

, W. B., Zero divisors in loop algebras of power

associative and di associative loops, Acta Politechnica Prace, CVUT, Praze,
16, 21-26 (1991).

57. V

ASANTHA

ANDASAMY

, W. B., Idempotents in rational loop algebras,

Ganit, 12, 53-56 (1992).

58. V

ASANTHA

ANDASAMY

, W. B., On normal elements in loop rings, U. Sci.

Phys. Sci., 4, 210-212 (1992).

59. V

ASANTHA

ANDASAMY

, W. B., Units in loop rings, U. Sci. Phys. Sci., 5, 86-

88 (1993).

143

60. V

ASANTHA

ANDASAMY

, W. B., On strictly right loop rings, J. Harbin Inst.

Sci. Tech., 18, 116-118 (1994).

61. V

ASANTHA

ANDASAMY

, W. B., Semi idempotents in loop algebra, J. Inst.

Math. Comp. Sci., 7, 241-244 (1994).

62. V

ASANTHA

ANDASAMY

, W. B., A note on the modular loop ring of a finite

loop, Opuscula Math., 15, 109-112 (1995).

63. V

ASANTHA

ANDASAMY

, W. B., On loop rings which are Marot loop rings,

U. Sci Phys. Sci., 7, 224-226 (1995).

64. V

ASANTHA

ANDASAMY

, W. B., On ordered groupoids and groupoid rings,

J. Math. and Comp. Sci., 9, 145-147, (1996).

65. V

ASANTHA

ANDASAMY

, W. B., Loop algebras of Moufang loops, Math

Forum, 11, 21-25 (1997).

66. V

ASANTHA

ANDASAMY

, W. B., On a new class of Jordan loops and their

loop rings, J. Bihar Math. Soc., 19, 71-75 (1999).

67. V

ASANTHA

ANDASAMY

, W. B., New classes of finite groupoids using Z

Varahmihir J. Math. Sci., 1, 135-143 (2001).

68. V

ASANTHA

ANDASAMY

, W. B., Smarandache zero divisors, (2001)

http://www.gallup.unm.edu/~smarandache/Zero-divisor.pdf

69. V

ASANTHA

ANDASAMY

, W. B., Smarandache semigroups, American

Research Press, Rehoboth, (2002).
http://www.gallup.unm.edu/~smarandache/Vasantha-Book1.pdf

70. V

ASANTHA

ANDASAMY

, W. B., Smarandache groupoids, American

Research Press, Rehoboth, (2002).
http://www.gallup.unm.edu/~smarandache/Vasantha-Book2.pdf

71. V

ASANTHA

ANDASAMY

, W. B., Smarandache Loops, Smarandache Notions

Journal, 13, 252-258 (2002).
http://www.gallup.unm.edu/~smarandache/Loops.pdf

72. V

ASANTHA

ANDASAMY

, W. B., Smarandache Loops, American Research

Press, Rehoboth, (2002).
http://www.gallup.unm.edu/~smarandache/Vasantha-Book4.pdf

144

73. V

ASANTHA

ANDASAMY

, W. B., On Smarandache Non-Associative rings,

(2002).
http://www.gallup.unm.edu/~smarandache/NA-rings.pdf

74. W

, Zhe-xian, Lie algebras (translated by Che Young Lee), Pergamon

press, (1975).

145

INDEX

Akisvis algebra, 123
Alternative (left), 14
Alternative (right), 14

Basis, 9, 110, 131
Bernstein algebra, 123
Bilinear form, 7-9
Bol groupoid, 61, 63
Bol loop, 14, 17
Bruck loop, 14

Centre, 54, 63, 124
Circle operation, 21, 23, 62
Closed net, 83, 102
Commutative groupoid, 57, 72, 77, 82
Composition algebra, 123, 125
Conjugate, 59
Conjugate pair, 59, 63, 69

Distributive lattice, 50

Eigen vectors, 10
E-ring, 69, 74-77
Essential ring, 90
Exponential ring, 87-88, 90
Exponentiation, 87-88

Finite dimensional, 8
F-ring, 83-84, 134
FZ-ring, 83

G-loop, 15
Group ring, 12, 53, 55, 61, 64, 76, 80
Groupoid, definition of, 11
Groupoid ring, definition of, 62

Hamiltonian loop, 44, 141
Hermitian form, 8

∗

-ring, 82, 99, 102

Ideal, 8
Infinite dimensional, 8
Inner commutative, 40, 48

Jacobi identity, 5, 97, 125
Jacobson radical, 22, 62
Jordan algebra, 77-79
Jordan groupoid, 58, 113
Jordan identity, 58, 109, 114-118, 122
Jordan loop, 111-112, 143
Jordan loop ring, 112
Jordan ring, 72, 112-113, 116
Jordan superalgebra, 123

KA-modules, 44
K-vectorial space, 8-9

Left conjugate, 58, 63
Left nucleus, 15
Leibniz algebra, 123-125
Lie algebra, definition of, 91
Lie ring, 72, 74, 96-99, 101-102, 104-

105, 135-136,

Lie superalgebra, 121, 123

146

Lin ring, 81, 133
Linear operator, 8, 10
Linear transformation, 8, 92, 95, 117
Linearly independent, 8
Loop, definition of, 11-12
Loop ring, definition of, 20

Marot loop ring, 40, 53
Marot ring, 40
Maximal solvable ideal, 94
Minimal solvable ideal, 94
Middle nucleus, 15
Minimal ideal, 94
Modular lattice, 30, 50-53, 80
Monoid, 11
Moufang center, 15
Moufang groupoid, 58

n-capacitor, 84, 88, 90, 105, 134
n-ideal groupoid ring, 66
Nil ideal, 88-89
Nilpotent ideal, 110
Normal groupoid, 61, 85
Normal groupoid ring, 63, 85, 90, 134
Normal subgroupoid, 58
Normal subring, 36, 69
Nucleus, 15, 131

Orthogonal ideal, 40, 45-46, 53, 130

P-groupoid, 58
pre-J-ring, 75-76
Principal isotopes, 15
p-ring, 74
Pseudo commutative, 19, 40, 49, 71,

76, 115

Q-groupoid ring, 82
Q-ring, 82-83
Quadratic Jordan algebra, 110

Quasi regular elements, 23-24, 29, 39,

40, 62, 128

Reduced, 73, 116
Regular ideal, 22, 39-41
Right alternative, 14, 27, 58, 139
Right conjugate, 59, 63
Right link relation, 89
Right nucleus, 15

Sagle algebra, 123-124
Semi-idempotent, 37-38
Semigroup, definition of, 7
Semigroup ring, 12, 64, 119
Semi nilpotent, 72, 115-116
Skew symmetric form, 7
Solvable Lie algebra, 94
Span v, 8
Strictly right commutative, 48, 130
Strongly modular lattice, 52-53, 131
Subgroupoid, 57-65
Supermodular lattice, 50-51, 53, 141-

142

Symmetric form, 7

Vector space, 7-10

Weak inverse property loop, 14
Weakly closed, 8
Weakly Q-ring, 82

Zassenhaus conjecture, 55-56, 132
Zero square ring, 69, 73-74

SMARANDACHE NOTIONS

S-Akivis algebra, 125
S-alternative groupoid, 61
S-associative triple, 19

147

S-basis, 9, 10, 101
S-Bernstein algebra, 124
S-Bol groupoid, 61, 69
S-Cartan subalgebra, 108, 137
S-Cauchy element, 18
S-characteristic value, 10
S-characteristic vector, 10
S-Chinese Lie ring, 102-103
S-clean element, 118, 137-138
S-closed net, 102
S-commutative groupoid, 60
S-commutator subloop, 19
S-conjugate Lie ring, 105
S-conjugate subgroupoid, 61
S-cyclic loop, 18
S-dispotent Jordan algebra, 117
S-E-Jordan algebra, 109
S-E-ring, 75-76
S-essential Lie algebra, 104
S-essential ring, 90
S-filial Lie ring, 101
S-finite dimensional filiform Leibniz

algebra, 124

S-generalized right semi-ideal, 68
S-groupoid, 7, 11, 59-61, 63-64, 68-

69, 78, 80, 84, 134

S-groupoid ring, 63-64, 69, 80
S-Hamiltonian loop, 44
S-ideal, 28, 37-38, 40-41, 50, 60, 66,

80-84, 89, 101-105, 131, 133, 137

S-idempotent, 31-32, 38, 40, 64, 69,

83, 102, 104, 117-118, 128

S-inner commutative, 49, 53, 61
S-inner commutative groupoid, 61
S-inner zero square ring, 74
S-Jacobson radical, 39
S-Jordan algebra, 113-118, 137
S-Jordan algebra homomorphism, 117
S-Jordan strong ideal property, 117
S-Jordan superalgebra, 123
S-K-vectorial space, 8, 9, 104
S-K-vectorial subspace, 9, 96, 100
S-Lagranges loop, 18
S-left ideal, 60
S-left semi-ideal, 67
S-Lie algebra, 91, 96, 99-108
S-Lie algebra homomorphism, 99
S-Lie ideal, 100-102, 107
S-Lie isomorphism, 99
S-Lie link related, 103

S-Lie semisimple, 107-108, 137
S-Lie simple, 107-108, 137
S-Lie subalgebra, 99
S-Lie superalgebra, 123
S-Lie super ore condition, 103-106
S-linear operator, 10
S-loop, 11-20, 25-26, 35, 43-44, 55
S-magnifying element, 117-118
S-mixed direct product type A, 91,

107-108

S-mixed direct product type B, 91,

107-108

S-mixed direct product type M, 125
S-mod p envelope, 43, 79
S-Moufang groupoid, 61
SNA-alternative ring, 28
SNA-ideal, 26-30, 127-128
SNA-left ideal, 26, 128
SNA-maximal ideal, 28
SNA-minimal ideal, 28
SNA-Moufang ring, 27
SNA-right ideal, 26
SNA-ring, 25-31, 55, 64, 71-75, 96,

127, 144

SNA-subring, 26-30, 75, 85, 96, 99,

127

SNA-units, 31, 39, 128
S-n-capacitor group, 88, 105
S-normal element, 36
S-normal groupoid, 60-61, 66, 85, 90
S-normal subloop, 17-18
S-normal subring, 66, 69
S-obedient ideal, 81, 84
S-orthogonal ideals, 46, 53
S-orthogonal Lie algebras, 108
S-pairwise associative loop, 19
S-P-groupoid, 61
S-power-joined Jordan algebra, 117
S-pre Boolean Jordan ring, 118
S-pre-J-ring, 75
S-prime pseudo ideal, 80
S-p-ring, 74-76
S-pseudo associative loop, 20
S-pseudo associator, 20
S-pseudo commutative loop, 19, 49
S-pseudo Jordan commutative pair,

115

S-pseudo zero divisor, 33-34
S-Q-ring, 83
S-quasicommutative Jordan algebra, 115

148

S-quasi regular element, 38, 40, 128
S-quotient ring, 82-83
S-radical ideal, 89
S-reduced, 73, 76
S-reduced Jordan algebra, 116
S-right commutative, 48
S-right ideal, 37, 60, 81, 128
S-right semi-ideal, 68
S-Sagle algebra, 124
S-semi conjugate subgroupoid, 60
S-semi-ideal, 68
S-semi idempotent, 38
S-seminormal groupoid, 60, 66
S-semigroup, 11, 88, 105, 118
S-shrinking element, 109
S-simple, 18, 61, 106-107, 137
S-strong basis, 10
S-strong Bol groupoid, 61
S-strong Jordan algebra, 114, 118
S-strong Moufang groupoid, 61
S-strong P-groupoid, 61
S-strongly commutative, 18-19, 65-66
S-strongly commutative loop, 18-19
S-strongly cyclic, 19
S-strongly Lie semi simple, 107, 137
S-strongly normal ring, 37
S-strongly right ideal ring, 37

S-strongly u.p loop, 44
S-subJordan algebra, 114
S-subgroup loop, 17
S-subgroupoid, 60-68, 71-80, 132
S-subloop, 17, 19-20, 43-44, 48-49,
129-130
S-subring link, 89
S-subring, 27, 36-37, 41, 48-50, 53,
64-66, 73-75, 81-83, 89-90, 101, 131
S-subsemi Lie ideal ring, 101
S-super ore condition, 104-105
S-sympletic Lie algebra, 108
S-system of local units, 118
S-triangular Lie algebra, 108
S-t.u.p loop, 44
S-u.p loop, 44, 130
S-uniformly power-joined, 117
S-unit, 31, 39, 54, 56, 69, 71, 118
S-vector space, 125
S-weak Marot ring, 41, 129
S-weak zero divisor, 35
S-weakly chain groupoid ring, 81
S-weakly G-Jordan ring, 118, 137
S-Zassenhaus conjecture, 55-56, 132
S-zero divisor, 33-35, 46, 69, 70
S-zero square ring, 73, 76

149

About the Author

Dr. W. B. Vasantha is an Associate Professor in the Department of Mathematics,
Indian Institute of Technology Madras, Chennai, where she lives with her husband
Dr. K. Kandasamy and daughters Meena and Kama. Her current interests include
Smarandache algebraic structures, fuzzy theory, neutrosophic theory, coding/
communication theory. In the past decade she has guided seven Ph.D. scholars in the
different fields of non-associative algebras, algebraic coding theory, transportation
theory, fuzzy groups, and applications of fuzzy theory of the problems faced in
chemical industries and cement industries. Currently, five Ph.D. scholars are
working under her guidance. She has to her credit 250 research papers of which 218
are individually authored. Apart from this she and her students have presented
around 230 papers in national and international conferences. She teaches both
undergraduate and post-graduate students and has guided over 35 M.Sc. and M.Tech
projects. She has worked in collaboration projects with the Indian Space Research
Organization and with the Tamil Nadu State AIDS Control Society.

She can be contacted at wbvasantha@indiainfo.com and vasantak@md3.vsnl.net.in

150

Definition:

Generally, in any human field, a Smarandache Structure on a set A means a
weak structure W on A such that there exists a proper subset B

⊂ A which is

embedded with a stronger structure S.

These types of structures occur in our everyday's life, that's why we study them
in this book.

Thus, as a particular case:

A Non-associative ring is a non-empty set R together with two binary operations
'+' and '.' such that (R, +) is an additive abelian group and (R, .) is a groupoid.
For all a, b, c

∈ R we have (a + b) . c = a . c + b . c and c . (a + b) = c . a + c . b.

A Smarandache non-associative ring is a non-associative ring (R, +, .) which has
a proper subset P

⊂ R, that is an associative ring (with respect to the same

binary operations on R).